Diagnostics Platform for Mitochondrial Dysfunctions/Diseases

Abstract

The present invention concerns machine learning based methods and systems for diagnosing and treating genetic diseases characterized by mitochondrial dysfunctions. A library of reference learning models is developed based on in vitro reference samples obtained from cell-cultures exposed to specific mitochondrial inhibitors. Each model is able to predict a specific labeled mitochondrial dysfunction induced in the cell-culture by the inhibitor/stressor. The reference models are then applied to target samples drawn in vivo from target subjects who are known to have specific genetic mitochondrial diseases. A mapping is developed between mitochondrial dysfunctions predicted in the subjects and their known mitochondrial diseases. This mapping and the reference models are then applied to a clinical sample of an undiagnosed patient in whom a diagnosis of a mitochondrial dysfunction and an associated mitochondrial disease is made. If there is a known rescuer for the mitochondrial dysfunction, it may be recommended in a personalized, targeted therapy.

Claims

1. A diagnostic method comprising the steps of: (a) introducing in one or more dosages a mitochondrial inhibitor into each of one or more cell-cultures grown in vitro from one or more cell-lines, said mitochondrial inhibitor inducing a mitochondrial dysfunction into said each of one or more cell-cultures; (b) drawing from each of said one or more cell-cultures one or more reference samples at one or more times since said introducing; (c) making one or more reference biomarker measurements from corresponding each of said one or more reference samples; (d) learning by a learning module one or more reference models each able to predict said mitochondrial dysfunction in an unseen biomarker measurement, said learning module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor; (e) drawing from one or more target subjects one or more target samples in vivo and making target biomarker measurements from corresponding said one or more target samples; (f) predicting by said one or more reference models said mitochondrial dysfunction in said one or more target subjects based on said one or more target biomarker measurements; and (g) matching by said targeting module said mitochondrial dysfunction to a mitochondrial disease known to exist in said one or more target subjects, said matching based on a statistically significant number of subjects from said one or more target subjects who are predicted to have said mitochondrial dysfunction in (f) above.

2. The method of claim 1 utilizing said one or more reference models in an ensemble to predict said mitochondrial dysfunction.

3. The method of claim 1 utilizing at least one of multiple linear regression and multiple logistic regression in said learning in (d) above.

4. The method of claim 1 utilizing a diagnosis module for applying said one or more reference models to a clinical biomarker measurement obtained from a clinical sample of an undiagnosed patient to predict said mitochondrial dysfunction and said mitochondrial disease in said undiagnosed patient, said diagnosis module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor.

5. The method of claim 4 based on a known rescuer for said mitochondrial dysfunction, providing for a personalized targeted therapy recommendation for said undiagnosed patient.

6. The method of claim 5 where said mitochondrial dysfunction causes one or more of a neurodegenerative disease, a cardiovascular disease, a type of diabetes, a metabolic syndrome, an autoimmune disease, a neurobehavioral disease, a psychiatric disease, a gastrointestinal disorder, a fatiguing illness, a musculoskeletal disease, a cancer and a chronic infection.

7. The method of claim 6 where said neurodegenerative disease comprises Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis (ALS) and Friedreich's ataxia.

8. The method of claim 6 where said cardiovascular disease is a vascular condition comprising atherosclerosis.

9. The method of claim 6 where said autoimmune disease comprises multiple sclerosis, systemic lupus erythematosus and Type 1 diabetes.

10. The method of claim 6 where said neurobehavioral disease comprises an autism spectrum disorder, schizophrenia, a bipolar disorder, a mood disorder, depression, attention deficit hyperactivity disorder (ADHD) and post-traumatic stress disorder (PTSD).

11. The method of claim 6 where said fatiguing illness comprises chronic fatigue syndrome and a Gulf War illness.

12. The method of claim 6 where said musculoskeletal disease comprises fibromyalgia and skeletal muscle atrophy.

13. A diagnostic platform comprising: (a) one or more reference models each able to predict a mitochondrial dysfunction in an unseen biomarker measurement made on a clinical sample obtained from an undiagnosed patient; (b) one or more cell-cultures grown in vitro from one or more cell-lines and said mitochondrial dysfunction induced in said one or more cell-cultures by an introduction in one or more dosages of a mitochondrial inhibitor; (c) said one or more reference models trained by a learning module based on reference biomarker measurements made from corresponding each of one or more reference samples drawn from said one or more cell-cultures at one or more times since said introduction, said learning module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor; (d) one or more target subjects in whom said mitochondrial dysfunction is predicted by said one or more reference models based on one or more target biomarker measurements made on corresponding one or more target samples drawn in vivo from said one or more target subjects; and (e) based on a statistically significant number of subjects from said one or more target subjects who are predicted to have said mitochondrial dysfunction in (d) above, an association developed by a targeting module between said mitochondrial dysfunction and a mitochondrial disease, said targeting module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor.

14. The platform of claim 13 further comprising a diagnosis module to predict by said one or more reference models said mitochondrial dysfunction in said undiagnosed patient based on said unseen biomarker measurement, and based on said association also said mitochondrial disease in said undiagnosed patient, said diagnosis module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor.

15. The platform of claim 14 further comprising a mass spectrometer to make one or more of said reference biomarker measurements, said target biomarker measurements and said unseen biomarker measurement.

16. The platform of claim 15 further comprising genomic data of said one or more target subjects obtained by one or more DNA sequencers and said mitochondrial disease is known to exist in said one or more target subjects based on said genomic data.

17. The platform of claim 13 wherein one or both of multiple linear regression and multiple logistic regression are used by said learning module.

18. The platform of claim 17 wherein a personalized targeted therapy for said undiagnosed patient is recommended based on said association of said mitochondrial dysfunction and said mitochondrial disease and on a known rescuer for said mitochondrial dysfunction.

19. The platform of claim 18 wherein said diagnosis module produces a diagnostic ranking for said undiagnosed patient, said diagnostic ranking containing a rank of said mitochondrial dysfunction, said mitochondrial disease, said mitochondrial inhibitor and a rescuer that is known to alleviate the effects of said mitochondrial dysfunction.

20. The platform of claim 18 wherein said mitochondrial dysfunction causes one or more of a neurodegenerative disease, a cardiovascular disease, a type of diabetes, a metabolic syndrome, an autoimmune disease, a neurobehavioral disease, a psychiatric disease, a gastrointestinal disorder, a fatiguing illness, a musculoskeletal disease, a cancer and a chronic infection.

21. The platform of claim 18 wherein said neurodegenerative disease comprises Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis (ALS) and Friedreich's ataxia.

22. The platform of claim 18 wherein said cardiovascular disease is a vascular condition comprising atherosclerosis.

23. The platform of claim 13 wherein said association exposes a correlation between said mitochondrial inhibitor and a genetic pattern of said undiagnosed patient.

24. A diagnostic system comprising: (a) one or more reference models each able to predict a mitochondrial dysfunction in an unseen biomarker measurement made on a clinical sample obtained from an undiagnosed patient; (b) one or more cell-cultures grown in vitro from one or more cell-lines and said mitochondrial dysfunction induced in said one or more cell-cultures by an introduction in one or more dosages of a mitochondrial inhibitor; (c) said one or more reference models trained by a learning module based on reference biomarker measurements made from corresponding each of one or more reference samples drawn from said one or more cell-cultures at one or more times since said introduction, said learning module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor; (d) one or more target subjects in whom said mitochondrial dysfunction is predicted by said one or more reference models based on one or more target biomarker measurements made on corresponding one or more target samples drawn in vivo from said one or more target subjects; (e) based on a statistically significant number of subjects from said one or more target subjects who are predicted to have said mitochondrial dysfunction in (d) above, an association developed by a targeting module between said mitochondrial dysfunction and a mitochondrial disease, said targeting module comprising a microprocessor executing program instructions stored in a non-transitory storage medium coupled to said microprocessor; (f) one or more mass spectrometers that are used to make at least one of said reference biomarker measurements, said target biomarker measurements and said unseen biomarker measurement; and (g) one or more DNA sequencing devices used to obtain sequenced genomic data of said one or more target subjects, and said mitochondrial disease is known to exist in said one or more target subjects based on said genomic data.

Description

BRIEF DESCRIPTION OF THE DRAWING FIGURES

[0059] FIG. 1 illustrates the diagnostic protocol according to the invention, comprising a learning phase, a targeting phase and a diagnosis phase.

[0060] FIG. 2 is a detailed system diagram illustrating its various components and interconnections for implementing the functionality of the present design.

[0061] FIG. 3 illustrates the typical lag, growth, stationary and death phases of a cell-culture.

[0062] FIG. 4 is an exemplary mass spectrum or a readout of the relative quantities of various analytes in a biomarker measurement of the present design.

[0063] FIG. 5 illustrates the piled-up construction of a 4.sup.th order input tensor of the instant learning framework based on the multidimensional input data of the present teachings.

[0064] FIG. 6 provides a more detailed view of the diagnostic protocol and further architectural details of the diagnostic platform of the instant technology.

DETAILED DESCRIPTION

[0065] The drawing figures and the following description relate to preferred embodiments of the present invention by way of illustration only. It should be noted that from the following discussion many alternative embodiments of the methods and systems disclosed herein will be readily recognized as viable options.

[0066] These may be employed without straying from the principles of the claimed invention. Likewise, the figures depict embodiments of the present invention for purposes of illustration only. Computer implemented learning methods and systems described herein will be best appreciated by initially reviewing the diagnostic protocol 50 as presented in FIG. 1. The diagnostic protocol of the present design has an initial learning phase as depicted by box 52. The learning phase is conducted on a number of in vitro samples that are obtained from cell-cultures grown from a number of cell-lines in a laboratory setting. As a result of this learning, a library of reference models 140A1, 140A2, . . . , 140AN are trained or learned or developed. Each of these models is able to predict a specific labeled mitochondrial dysfunction or dysfunctions given measurements of biological markers obtained from a biological sample. Each trained reference model 140A1, 140A2, . . . , 140AN is able to detect the biomarker fingerprint/signature that is indicative of a specific mitochondrial dysfunction.

[0067] The next phase of protocol 50 is referred to as targeting as shown by box 54. In targeting phase 54, models 140A1-AN are used to predict labeled mitochondrial dysfunction(s) in samples obtained from in vivo patients who have known genetic mitochondrial diseases based on genetic defects observed in their sequenced genomic data. More generally, these patients are known to have genetic diseases that are characterized by mitochondrial dysfunctions and which may or may not be expressed by the genetic information observed in the sequenced genetic data. In any case, as a resulting of targeting 54, a mapping or correspondence or association 130 of mitochondrial dysfunctions predicted by models 104A1-AN and corresponding genetic mitochondrial diseases is obtained.

[0068] Finally, in clinical, field or diagnosis phase 56 of our diagnostic protocol 50, a clinical biomarker measurement from a sample of an undiagnosed patient is used to predict the presence of a mitochondrial dysfunction(s) and any associated genetic mitochondrial diseases in that patient. The presence of any genetic diseases is determined based on mapping 130 obtained during targeting 54 and not on any genomic data obtained and sequenced from the patient. Then, based on mapping 130, diagnostic protocol 50 of the present design may be used to recommend new and personalized therapies to the patient that were heretofore unknown.

[0069] Let us now study each phase of protocol 50 of FIG. 1 in great detail. For this purpose, let us take advantage of the system diagram of the present technology as illustrated in FIG. 2. This diagram shows the key parts and interconnections of a diagnostic platform or system 100 configured to diagnose mitochondrial dysfunctions in biological processes or bioprocesses. The bioprocesses are being experienced by reference in vitro biological entities or reference in vitro cultures obtained from biological entities. Unless otherwise noted, the terms biological entities and cell-cultures that were grown from biological entities are used interchangeably in this disclosure and are referenced by numerals 102A, 102B, . . . , 102Z in FIG. 2. These biological entities or cell-cultures or simply cultures reside in respective bioreactors 104A, 104B, . . . , 104Z as shown.

[0070] It should be noted that any number of such in vitro biological cultures 102A-Z in respective bioreactors 104A-Z may be present and unless otherwise noted, the reference numerals used in this example or other examples in this disclosure are non-limiting. As such, reference numerals such as 102A-Z, 152A-X, etc. as used throughout these teachings are understood to mean any number of elements 102 rather than just 26 (A through Z) and any number of elements 152 rather than just 24 (A through X).

[0071] In the embodiment shown in FIG. 2, reference in vitro cultures 102A-Z are one or more biomasses, cell-cultures, biomaterials or biologically active substances undergoing the bioprocesses of interest as will be described below. Bioreactors 104A-Z should be understood to include dedicated reactors as well as incidental mechanisms. These include any manufactured or engineered device or system that supports a biologically active environment for growing cells or tissues, including petri dishes or cell-culture dishes. Thus, reference conditions experienced by reference in vitro cultures or biological entities 102A-Z are those existing or sustained inside bioreactors 104A-Z respectively.

[0072] At their broadest level, bioreactors 104A-Z include an engineered or managed system that supports a biologically active environment in which a chemical process is carried out that involves biological organisms or biochemically active substances or in vitro cultures 102A-Z. Bioreactors 104A-Z presented herein may range from small scale bioreactors, on the order of 10s to 100s of mL, to larger scale reactors of thousands or tens of thousands of liters. In particular, the bioreactors will typically have a volume of greater than 10 mL, greater than 100 mL, greater than 500 mL, greater than 1 L, greater than 5 L, greater than 10 L, greater than 100 L, greater than 500 L, greater than 1000 L, greater than 5000 L, greater than 10,000 L, greater than 100,000 L.

[0073] Bioprocesses of interest in the present invention involve those that include reduction-oxidation reactions. The energy involved in such a bioprocess is indicated by the voltage or potential difference V equal to the redox potential E.sub.h. The exact numeric value of redox potential E.sub.h will depend on departure of thermodynamic conditions from standard conditions, as described by the well-known Nernst equation E.sub.h=E.sub.o+RT/nF.Math.ln([A]/[B]). Here E.sub.o is the standard potential for the redox couple, R is the ideal gas constant, T is the absolute temperature in degrees Kelvin, n is the number of electrons transferred in the redox reaction and F is Faraday's constant. We use the natural logarithm of the ratio of concentrations (indicated by square brackets) of the oxidized and reduced members of the redox couple A, B (e.g., NAD+ and NADH, glutathione couple GSH/GSSH or cysteine and cystine couple Cys/CySS). Those skilled in the art will also be aware of still other parameters and factors that need to be considered in assessing the redox potential of any particular redox couple (e.g., whether it is in cell, in plasma, etc.).

[0074] The redox status of a large number (e.g., hundreds or thousands) of redox couples is measurable, especially under lab conditions. On large scales, electron balance induces changes in well-known parameters, e.g., the pH value (which is a common measure of H.sup.+ ion concentration in moles per liter of solution expressed on a logarithmic scale). Persons skilled in the art will be very familiar with measurements of redox status using such parameters. These parameters are commonly referred to as electron balance indicators or redox indicators. Depending on conditions and available equipment, the most useful group of redox indicators can include certain oxidoreductases, oxidoreductase co-factors, electron balance influencer compounds, electron balance influencer compositions, redox-active compounds, pK values, pH values, threshold values, context measures and soft or derived indicators (usually derived with reference to a mathematical model).

[0075] FIG. 2 also shows a general apparatus used by diagnostic system 100 to learn, measure and control or adjust the redox status of the bioprocesses that reference in vitro cultures 102A, 102B, . . . , 102Z are undergoing. Respective inputs 106A, 106B, . . . , 106Z to reference bioreactors 104A-Z are provided for adjusting or altering the bioprocesses occurring inside them. Inputs 106A-Z are generally to be understood as any mechanism, actuator, inlet or other type of mechanical or non-mechanical apparatuses capable of acting on the bioprocess. Likewise, output sensors 108A, 108B, . . . , 108Z are provided for making measurements on outputs or samples 103A, 103B, . . . , 103Z drawn from the bioprocesses unfolding inside reference bioreactors 104A, 104B, . . . , 104Z respectively. These output sensors may be any type of sensing devices including mass spectrometers or even soft sensors with readings derived from various other measurements or proxies.

[0076] According to the main aspects of the present disclosure, a learner or data processing module 120 is used to learn any number of reference models 140A1, 140A2, . . . , 140AN. Each of models 140A1-AN is capable of predicting a mitochondrial dysfunction or dysfunctions induced in a biological entity or cell-culture due to a specific mitochondrial inhibitor, inducer, stressor or insult introduced into the entity or culture. Typically, the mitochondrial inhibitor is introduced at the beginning of the stationary phase of cell-growth of the culture although it can be done at any time during the life of the culture. This will be further discussed in reference to FIG. 3 in this disclosure.

[0077] This prediction is based on a biomarker measurement obtained from a sample drawn/taken from the entity or culture. For this purpose, learner 120 deploys one or more machine learning algorithms for learning reference models 140A1-AN. Learner 120 runs on a dedicated computer, computer system or even a computer cluster that is collocated or geographically distributed (not shown). A person skilled in the art will appreciate, that many types of resources and architectures can support the execution of data processing module or learner 120. Furthermore, processing module 120 is understood to execute program instructions by one or more processors in order to carry out its functions as described herein. The program instructions are stored in one or more non-transitory storage media that is/are coupled to the one or more microprocessors.

[0078] Output sensors 108A-Z perform reference measurements on reference in vitro samples taken from in vitro cultures 102A-Z. The reference in vitro samples taken for measurements are shown by vials of which only one is marked by reference numerals 103A, 103B, . . . , 103Z to avoid clutter. In practice, instead of or in addition to vials, any other mechanical or non-mechanical biological sampling mechanisms including tubing, suction and other techniques may also be used to extract samples 103A-Z.

[0079] Once a sample is drawn/extracted at a given instant of time, it is sealed to stop further reaction and oxidation. A sample, for example one of samples 103B, is thus representative of the stage or the moment in time of the reaction of culture 102B at which that specific sample 103B was drawn. Recall from above that reference numeral 103B may represent more than one samples extracted from culture 102B. The techniques for drawing/extracting samples 103A-Z from cell-cultures or biological entities 102A-Z are known in the art and will not be delved into detail in this disclosure.

[0080] The reference measurements result in respective biomarker measurements 110A-Z of reference in vitro cultures 102A-Z, each measurement corresponding to an instant of time or a stage of the reaction of the culture at which the respective sample 103A-Z was drawn. Thus, each of in vitro samples 103A will result in a reference biomarker measurement, however only one such biomarker measurement is marked with reference numeral 110A for reasons of clarity. For the same reasons, only one of biomarker measurement obtained from samples 103B is marked with 110B and so on. Reference biomarker measurements 110A-Z contain measured quantities of analytes belonging to several categories of redox data based on the redox code. The measurements may contain analytes or metabolites belonging to the various metabolomes related to the bioprocesses which cultures 102A-Z are undergoing.

[0081] The redox code includes the four principles by which biological systems are organized. The first category contains bio-energetics redox data. These are data pertaining to catabolism and anabolism typically organized through high-flux NAD and NADP systems. The second category contains macromolecular structure and activities that are linked to bio-energetic systems through kinetically controlled sulfur switches. This category is referred to as switching redox data. The third category contains signaling redox data. This category relates to activation and deactivation cycles, e.g., of H.sub.2O.sub.2 production (usually linked to NAD and NADP systems to support redox signaling and spatiotemporal sequencing for differentiation and multicellular development). The fourth category contains network redox data. This type of data relates to redox networks, from micro-compartments to subcellular and cellular organization and includes adaptive responses to the environment.

[0082] In addition to the four redox code categories, reference biomarker measurements 110A-Z may also contain a fifth category of data. This fifth category includes contingent redox data. Contingent redox data includes candidates (e.g., candidate redox indicators that are speculative) for any of the first four categories, as well as contextual information having to do with reference conditions or environment in which reference bioprocess of in vitro cell-cultures 102A-Z transpire. Contingent data can also include other types of information that may be relevant directly or indirectly to oxidation-reduction activity or charge balance. It is possible for contingent redox data to encompass contextual information that can only be inferred from factors not specifically related in any known way to charge balance. Contingent redox data can also include common annotations, labels and other information that curators or experts typically add to ensure proper understanding of the data.

[0083] Reference biomarker measurements 110A-Z can also include information that is not directly measurable, also known herein as soft data. Such soft data is inferred on a model applied to a collection of surrogate measures that are weighted to estimate or infer a measure of interest. For more information about soft sensors and soft data the reader is referred to Paulsson D., et al., A Sensor for Bioprocess Control Based on Sequential Filtering of Metabolic Heat Signals, Vol. 14, Sensors, 26 Sep. 2014, pp. 17864-17882.

[0084] Processing module 120 is configured to receive reference biomarker measurements 110A-Z from reference in vitro samples taken from respective in vitro biological entities or cell-cultures 102A-Z grown in the lab. In the event that biological entities or in vitro cultures 102A-Z undergoing the bioprocesses in reference bioreactors 104A-Z require frequent or even continuous monitoring, the delay in the communication of reference biomarker measurements 110A-Z to learner 120 should be kept as short as practicable. In such cases, geographic collocation of the computer(s) running processing module 120 with bioreactors 104A-Z containing in vitro samples 102A-Z is preferred. A person skilled in the art will be able to make the appropriate decision about the distribution and assignment of the correspondent computational tasks and in vitro samples 102A-Z in the lab conditions.

[0085] In accordance with the instant design, processing module 120 deploys one or more machine learning algorithms for learning from reference biomarker measurements 110A-Z, reference models 140A1-AN. The learning algorithms may operate in supervised or unsupervised mode. After the learning or training, each reference model 140A1-AN is able to predict a labeled mitochondrial dysfunction present in any biological sample based on its biomarker fingerprint detected in input biomarker measurements 110A-Z. In other words, the biomarker fingerprint itself is the specific subset or pattern or principal components/weights from all measurements 110A-Z in the lab which most successfully predicts D1. We also at times refer to such a fingerprint as conserved enough across the runs to be able to predict D1.

[0086] In this disclosure, the set of dysfunction/dysfunctions thus predicted are identified by a label, such as D1, D2, etc. each associated with the mitochondrial inhibitor, stress or insult known to induce the mitochondrial dysfunction. In other words, the training data for the machine learning process is labeled by the known dysfunction associated with the sample from which the data was generated. More specifically, data processing module or processing module or learning module or simply learner 120 trains a number of machine learning models 140A1-N that predict the presence of a certain mitochondrial dysfunction/dysfunctions present in the biological sample from which the biomarker measurement was obtained. This learning process will be taught in much more detail further below.

[0087] Learner 120 learns reference models 140A1-AN on training data based on reference biomarker measurements obtained from in vitro samples drawn from in vitro cell-cultures 102A-Z that are perturbed by respective inputs or actuator mechanisms 106A-Z operating on them. Actuator systems 106A, 106B, . . . , 106Z interface with respective bioreactors 104A, 104B, . . . , 104Z. Each of actuators 106A-Z deploys one or more individual input mechanisms to control, provide inputs to, or in any other way, alter or perturb or adjust the bioprocess transpiring in the respective reference biological entity or reference in vitro culture 102A-Z housed in respective bioreactor 102A-Z.

[0088] It is noted that in some embodiments any of actuators 106A-Z may only utilize on or more actuator or input mechanisms, e.g., a stirrer or just an inlet pipe or multiple inputs or inlet pipes, coupled to multiple sources of inputs to supply additional quantities of culture material to in vitro cultures 102A-Z, or to provide still any other input material. These other inputs could include other feed stock or biomaterials, including, e.g., redox influencers or mitochondrial inhibitors, inducers, stressors or insults. Further, the inputs may also include: off-gas, air, O.sub.2, CO.sub.2, pressure, viscosity, stirrer speed, temperature, pO.sub.2, pH, photometrics, calorespirometric measures and other biomeasureables. Of course, there may be cases in which control of the local bioprocess is impossible or impractical. This could occur in rapidly transpiring reactions or reactions that go to completion without allowing for meaningful intervention. Actuator system 106 can also recommend an operation to a local operator (not shown).

[0089] Specific bioprocesses transpiring in each reference biological entity or in vitro culture 102A, 102B, . . . , 102Z are sensed or measured at various stages by first drawing in vitro samples 103A-Z from the cultures and then monitoring and/or measuring them by corresponding sensor systems 108A, 108B, . . . , 108Z. Although not explicitly shown, each sensor system 108A, 108B, . . . , 108Z may include one or more individual measurement devices, sensors and/or monitors as well as any requisite interfaces, hardware and software.

[0090] Sensors 108A-Z can also include high-resolution and high-throughput mass spectrometers. Reference biomarker measurements 110A-Z can take into account mass spectrometer results resolving as many as 20,000 or even 50,000 or more potential peaks to locate known or targeted redox indicators for the bioprocesses of in vitro samples 102A-Z. Alternatively, if sufficient processing power is employed, a full or partially untargeted set of peaks may be measured to associate the mitochondrial dysfunction label with patterns of analytes without any previous knowledge about the particular mechanisms of the analytes.

[0091] The above is advantageously accomplished by using a high-resolution mass spectrometer in which mass-to-charge ratio (m/z) for each ion is measured to several decimal places to differentiate between molecular formulas having similar masses. Potential mass spectrometers include instruments supplied by commercial manufacturers such as AB Sciex, Advion, Agilent, Applied biosystems, Bruker, GenTech Scientific, Hitachi High Technologies, IONICON, JEOL, LECO, PerkinElmer, Shimadzu, Thermo Fisher Scientific, Waters and others.

[0092] Measurements 110A-Z used to train each of reference models 140A1-AN to predict a labeled mitochondrial dysfunction can comprise a range of mass spectrometry outputs/peaks, any other sensor data for specific analyte measurements or even the conditions of bioreactors 104A-Z, any soft sensors of the bioreactors, or still any other type of measurement data. Respective inputs/actuators 106A-Z can be used and other settings or environmental conditions of the bioreactors can be varied to create more data sets so that the models predicting the mitochondrial dysfunctions are more stable over a broad range of conditions. This way, the models are not over-trained on data/conditions not conserved across the wide variety of conditions/situations characteristic of mitochondrial dysfunction diagnostics under in vivo conditions.

[0093] In many practical situations, a mass spectrometer is a valuable and shared resource, so it will be prudent that in vitro samples 103A-Z taken from multiple respective biological entities or cultures 102A-Z are measured by a common mass spectrometer. As such, mass spectrometer or sensor 108 will be a device common for measuring the outputs of the one or more cultures. This sharing of the same sensor/spectrometer 108 in FIG. 2 is illustrated by dotted line 109. For this reason, it is also desirable to have sensor/spectrometer 108 that is high-throughput and is able to process a large volume of samples at preferably high-resolution with efficiency.

[0094] Diagnostic platform 100 of FIG. 2 is understood to include the requisite control interfaces operatively coupled to actuators 106A-Z as well as the requisite processors and/or control interfaces coupled to sensors 108A-Z. These processors and/or control interfaces may be instructed and supervised by processing module 120 according to the techniques provided herein. Alternatively, they may be controlled by other external systems and devices not specifically shown in FIG. 2.

[0095] For example, once samples 103A-Z are extracted from cultures 102A-Z respectively at the desired times, respective sensors 108A-Z may obtain biomarker measurements at their own behest or under the supervision of an external module and provide or push the corresponding data in the form of respective measurements 110A-Z to processing module 120. Alternatively, processing module 120 may instruct sensors 108A-Z to perform their reference measurements and subsequently pull the measurement data in the form of respective biomarker measurements 110A-Z.

[0096] In the preferred embodiment, each of in vitro cultures 102A-Z is grown in the lab from a specific cell-line. In a given run or a set of experiments, there may be several cultures of a given cell-line present. Then, reference actuator mechanisms 106A-Z introduce a given/selected mitochondrial inhibitor/inducers/stressor/insult into each of in vitro cultures 102. This introduction is preferably performed at the start of the stationary phase of cell-growth of the respective cultures, or simply stated the stationary phase of the respective cultures.

[0097] Furthermore, the introduction is preferably carried out in varying dosages. In other words, the cultures from a given cell-line may each be given a varying dosage of the mitochondrial inhibitor so that their biomarker measurements may be made under different exposure concentrations.

[0098] One of the dosage amounts needs to be 0/zero or in other words, no introduction of the mitochondrial inhibitor. This 0/zero state is necessary because the learning process needs to recognize those patterns that are just generated by a given in vitro culture in its natural or uninhibited state. It will then need to eliminate those patterns when predicting mitochondrial dysfunction in the culture in when it is in its perturbed state.

[0099] Alternatively, the zero state may be thought of as the labeled state defined by a null hypothesis inhibitor. In such a scenario, learning module/engine 120 can be used to train a model that predicts the zero state just as it can train a model to predict the labeled mitochondrial dysfunction caused by a specific mitochondrial inhibitor. In either case, the label corresponds to a state of mitochondrial function/dysfunction that is measured in biomarker measurements 110A-Z in vitro. The learning process will be explained in detail further below.

[0100] Reference measurements are also preferably performed a number of times in a time series manner by sensor mechanisms 108A-Z on respective reference in vitro cultures 102A-Z. Explained further, each of these reference measurements are performed on respective in vitro samples 103A-Z drawn from cultures 102A-Z at times t.sub.1, t.sub.2, . . . , t.sub.i. Typically, times t.sub.1, t.sub.2, . . . , t.sub.i are measured as time intervals since the exposure of the culture to the mitochondrial inhibitor chosen for the runs. Recall that the culture is preferably exposed to the mitochondrial inhibitor at the start of the stationary phase of the cell-growth of the culture, although this exposure may be done at any time during the life of the culture.

[0101] Thus, reference numeral 103A refers to one or more samples drawn from culture 102A at varying times t.sub.1, t.sub.2, . . . , t.sub.i since the exposure/perturbation of the culture with the specific mitochondrial inhibitor. Similarly, reference numeral 103B refers to one or more samples drawn from culture 102B at varying times t.sub.1, t.sub.2, . . . , t.sub.i since the exposure/perturbation of the culture with the same specific mitochondrial inhibitor, and so on. In some embodiments, samples 103 may also be drawn from cultures 102 under differing environmental conditions that may influence the cell culture, such as temperature, pressure, exposure to gases, etc. Such differing conditions may be effected by actuators/inputs 106 as noted above.

[0102] A measurement of biomarkers or analytes from a given in vitro sample is referred to as a reference biomarker measurement 110. Thus, reference numeral 110A refers to one or more reference biomarker measurements from respective samples 103A at varying times t.sub.1, t.sub.2, . . . , t.sub.i since the exposure/perturbation of culture 102A with the mitochondrial inhibitor and under varying conditions. Similarly, reference numeral 110B refers to one or more reference biomarker measurements from respective samples 103B at varying times t.sub.1, t.sub.2, . . . , t.sub.i since the exposure/perturbation of culture 102B with the mitochondrial inhibitor and under varying conditions, and so on. Note that there is no requirement that the number of samples 103A-Z drawn from respective cultures 102A-Z resulting in corresponding biomarker measurements 110A-Z be all the same in number. The equal number (3) of vials of samples 103A, 103B, . . . , 103Z and biomarker measurements 110A, 110B, . . . , 110Z shown in FIG. 2 is for exemplary purposes only.

[0103] For a set of runs involving a given mitochondrial inhibitor applied to a given number of cell-cultures grown from a given number of cell-lines, the total number of reference measurements will be:

(1 mitochondrial inhibitor+1 (when no mitochondrial inhibitor is used))number of cell-cultures grownnumber of cell-linesi (number of time series measurements)).

[0104] Each of these reference measurements is measured by sensor 108A-Z, such as a high-resolution, high-throughput mass spectrometer.

[0105] In a similar manner, another set of reference runs or experiments is conducted for a different mitochondrial inhibitor, and so on for other mitochondrial inhibitors.

[0106] The total number of reference in vitro measurements are thus:

(Total number of mitochondrial inhibitors+1)Total number of cell-cultures from each cell-lineTotal number of cell-linesSampling rate i or the total number of times at which the measurements are taken.

[0107] In an exemplary scenario, with 13 inhibitors and 20 cell-cultures from 97 cell-lines and with 20 time-based measurements, the total number of reference biomarker measurements are: 14972020=543200.

[0108] Table 1 lists exemplary neurobiology cell-lines that may be used for growing in vitro cell-cultures 102A-Z while Table 2 lists exemplary mitochondrial inhibitors, both provided by Sigma-Aldrich. Thus, exemplary inhibitors/inducers/stressors/insults in Table 2 may be used as perturbations for exemplary cultures derived from Table 1 in the above described reference runs/experiments.

TABLE-US-00001 TABLE 1 No. Cell Name Description 1 DI TNC1 Rat Astrocyte transfected 2 CTX TNA2 Rat Astrocyte, Transfected 3 BE(2)-C Human Caucasian neuroblastoma 4 BE(2)-M17 Human Caucasian neuroblastoma 5 SK-N-BE(2) Human Caucasian neuroblastoma 6 SK-N-DZ Human neuroblastoma 7 SH-SY5Y Human neuroblastoma 8 C1300 Mouse neuroblastoma CLONE NA 9 ND-C Mouse neuroblastoma Rat neurone hybrid 10 ND27 Mouse neuroblastoma Rat neurone hybrid 11 ND15 Mouse neuroblastoma Rat neurone hybrid 12 ND8/34 Mouse neuroblastoma Rat neurone hybrid 13 ND7/23 Mouse neuroblastoma Rat neurone hybrid 14 ND3 Mouse neuroblastoma Rat neurone hybrid 15 C6 Rat glial tumour 16 NB4 1A3 Mouse C-1300 Neuroblastoma 17 Neuro 2a Mouse Albino neuroblastoma 18 N1E-115 Mouse neuroblastoma 19 NG108-15 Mouse neuroblastoma Rat glioma hybrid 20 N18 Mouse neuroblastoma Rat glioma hybrid 21 33B Rat nervous tissue oligodendroglioma 22 B65 Rat nervous tissue neuronal 23 B92 Rat nervous tissue glial 24 B50 Rat nervous tissue neuronal 25 CAD Mouse (B6/D2 F1 hybrid) catecholaminergic neuronal tumour 26 F11 Rat embryonic dorsal root ganglion 27 108CC5T-BU-4 Mouse neuroblastoma Rat glioma hybrid 28 108CC5T-BU-1 Mouse neuroblastoma Rat glioma hybrid 29 108CC5-BU-8 Mouse neuroblastoma Rat glioma hybrid 30 108CC5-BU-5 Mouse neuroblastoma Rat glioma hybrid 31 108CC5-BU Mouse neuroblastoma Rat glioma hybrid 32 N115-BU-10 Mouse Neuroblastoma 33 N115-BU-9 Mouse Neuroblastoma 34 N115-BU-7 Mouse Neuroblastoma 35 N115-BU-2 Mouse neuroblastoma, bromodeoxyuridine resistant 36 108CC5-TG-4 Mouse neuroblastoma Rat glioma hybrid 37 108CC5-TG-3 Mouse neuroblastoma Rat glioma hybrid 38 108CC5-TG-2 Mouse neuroblastoma Rat glioma hybrid 39 108CC5-TG-1 Mouse neuroblastoma Rat glioma hybrid 40 328/14 Mouse neuroblastoma mouse L cell fibroblast hybrid 41 328/12 Mouse neuroblastoma mouse L cell fibroblast hybrid 42 328/11 Mouse neuroblastoma mouse L cell fibroblast hybrid 43 328/10 Mouse neuroblastoma mouse L cell fibroblast hybrid 44 328/9 Mouse neuroblastoma mouse L cell fibroblast hybrid 45 328/8 Mouse neuroblastoma mouse L cell fibroblast hybrid 46 328/7 Mouse neuroblastoma mouse L cell fibroblast hybrid 47 NH15-CA2 Mouse neuroblastoma Rat glioma hybrid 48 N18TG2 Mouse neuroblastoma 49 108CC5 Mouse neuroblastoma Rat glioma hybrid 50 N115-BU-8 Mouse Neuroblastoma 51 NS20Y Mouse neuroblastoma 52 108CC15 Mouse neuroblastoma Rat glioma hybrid 53 N4TG3 Mouse neuroblastoma 54 NS20Y-TG Mouse neuroblastoma 55 N1E-115-1 Mouse neuroblastoma 56 NS20Y-BU-7 Mouse neuroblastoma 57 NS20Y-BU-6 Mouse neuroblastoma 58 NS20Y-BU-5 Mouse neuroblastoma 59 NS20Y-BU-4 Mouse neuroblastoma 60 NS20Y-BU-2 Mouse neuroblastoma 61 C17.2 Mouse multipotent neural progenitor or stem-like cells 62 CHP-134 Human neuroblastoma tumour mass of left adrenal gland 63 LAl-5s Human Neural Crest-Derived Non- Neuronal Progenitor 64 LA-N-1 Human Neuroblastoma Bone Marrow Metastasis 65 WERI Human Retinoblastoma 66 Rolf B1.T Adult rat olfactory nerve ensheathing cells 67 Y79 Human Caucasian retinoblastoma 68 RB247C Human Retinoblastoma 69 1321N1 Human brain astrocytoma 70 A15 Rat, BDIX, glioma 71 ANGM-CSS Human glioblastoma 72 b.End3 Mouse SV129 brain endothelioma 73 b.End5 Mouse Balb/c brain endothelioma 74 BC3H1 Mouse brain tumour 75 BE10-7 Rat, BDIX, brain, pre-malignant 76 BE10- Rat, BDIX, foetal brain, pre- Intermediate malignant 77 BE10-Late Rat, BDIX, foetal brain, malignant 78 BE11 Rat, BDIX, foetal brain (Early) 79 C6-2-3 Rat glioma rat glioma hybrid 80 C6-4-2 Rat glioma rat glioma hybrid 81 C6-BU-1 Rat glioma rat glioma hybrid 82 CCF-STTG1 Human Caucasian astrocytoma 83 DBTRG.05MG Human glioblastoma 84 GPNT An immortalised Lewis rat brain vascular endothelial cell-line 85 KELLY Human neuroblastoma 86 MOG-G-CCM Human brain astrocytoma 87 MOG-G-UVW Human brain astrocytoma 88 NB69 Human neuroblastoma (Stage III) 89 PG-4 Cat brain Moloney sarcoma virus- transformed 90 SCP Ovine brain choroid plexus 91 T98G Human Caucasian glioblastoma 92 TR33B Rat Wistar-Furth oligodendroglioma 93 U-87 MG Human glioblastoma astrocytoma 94 U-251 MG Human glioblastoma astrocytoma (formerly known as U-373 MG) 95 U-373 MG Human glioblastoma astrocytoma (Uppsala) 96 RN46A-B14 Embryonic rat medullary raphe, temperature-sensitive mutant of SV40 large T-antigen, immortalised, serotonergic, neuronal 97 RN4 6A Embryonic rat medullary raphe, temperature-sensitive mutant of SV40 large T-antigen, immortalised, serotonergic, neuronal

TABLE-US-00002 TABLE 2 Mitochondrial Mitochondrial dysfunction Label Inhibitor Description induced D1 A8674 Antimycin A Inhibits electron transfer at from complex III. Induces apoptosis. Streptomyces sp. D2 BM0017 BMS-199264 Potently inhibits the ATP hydrochloride 98% hydrolase activity of (HPLC) mitochondrial FIFO ATP synthase. The compound BMS- 199624 has no effect on the ATP synthase function of FIFO. In isolated rat hearts, BMS-199624 blocks depletion of ATP levels, and blocks necrosis during ischemia. D3 SML1122 BTB06584 98% BTB06584 inhibits the ATP (HPLC) hydrolase activity of mitochondrial FIFO ATP synthase. The compound BTB06584 has no effect on oxygen consumption or mitochondrial membrane potential in HL-1, a mouse cardiac cell-line, but blocks ATP consumption and ischemic cell death following inhibition of cellular respiration. D4 C2759 Carbonyl Protonophore (H+ ionophore) and cyanide 3- uncoupler of oxidative chlorophenyl- phosphorylation in hydrazone 97% mitochondria. Shown to have a (TLC), powder number of effects on cellular calcium. Inhibits secretion of hepatic lipase and partially inhibits the pH gradient- activated Cl- uptake and Cl-/ Cl- exchange activities in brush-border membrane vesicles. D5 C2920 Carbonyl FCCP is a protonophore (H+ cyanide 4- ionophore) and uncoupler of (trifluoro- oxidative phosphorylation in methoxy)phenyl- mitochondria. It is capable of hydrazone 98% depolarizing plasma and (TLC), powder mitochondrial membranes. FCCP has been shown to have a number of effects on cellular calcium. It also is reported to inhibit a background K+ current and induce a small inward current, reduce pH by 0.1 unit, and induce a rise of intracellular [Na+]. FCCP stimulates Mg2+- ATPase activity, inhibits - amyloid production, and mimics the effect of selective glutamate agonist N-methyl-D- aspartate (NMDA) on mitochondrial superoxide production. D6 C2020 -Cyano-4- Specific inhibitor of hydroxycinnamic monocarboxylic acid transport, acid 98% including lactate and pyruvate (TLC), powder transport. Also reported to block -cell apical anion exchange (IC50 of 2.4 mM). D7 I9890 m-Iodobenzyl-guanidine Antitumor agent which inhibits hemisulfate ADP ribosylation. salt 98% (HPLC and TLC) D8 L4900 Lonidamine Inhibits the energy metabolism mitochondrial of neoplastic cells by hexokinase interfering with hexokinase and inhibitor disrupting uncoupler-stimulated mitochondrial electron transport; damages cell and mitochondrial membranes. D9 M2324 ML-3H2 ML-3H2 is an allosteric hexamer-stabilizing inhibitor of human porphobilinogen synthase (PBGS; ALAD) D10 04876 Oligomycin Macrolide antibiotic; inhibits from mitochondrial ATPase and Streptomyces phosphoryl group transfer. diastato- chromogenes 90% total oligomycins basis (HPLC) D11 P8861 Pyrrolnitrin Pyrrolnitrin blocks the from terminal electron transport Pseudomonas between succinate or reduced cepacia 98% NADH and coenzyme Q. In (HPLC), solid mitochondria preparations of S. cerevisiae, the antibiotic inhibited succinate oxidase, NADH oxidase, succinate cythochrome C reductase, and NADH-cytochrome C reductase. Pyrrolnitrin is involved in many cellular processes such as oxidative stress, electron transport, DNA and RNA synthesis. D12 R8875 Rotenone 95% Inhibitor of mitochondrial electron transport at NADH:ubiquinone oxidoreductase. It is readily absorbed through the exoskeletons of arthropods, but poorly absorbed cutaneously or from the gastrointestinal tract of mammals. Rotenone is used to induce a Parkinson-like syndrome as an experimental model in rats. Inhibitor of mitochondrial electron transport. Neurotoxic agent that can produce a Parkinson-like condition as an animal model for study of etiology and interventions. D13 SML1280 TT01001 98% TT01001 is potent, orally (HPLC) available the mitochondrial outer membrane protein mitoNEET ligands that binds to mitoNEET without PPAR activation. TT01001 improves hyperglycemia, hyperlipidemia, and glucose intolerance in mice models of diabetes II. TT01001 exerts anti-diabetic effects without the pioglitazone associated weight gain.

[0109] Each biomarker measurement 110A-Z consists of the measurements indicating the presence and concentrations of a variety of analytes observed in the sample by an appropriate sensor, such as a mass spectrometer. In this disclosure, analytes refer to any element/compound such as a metabolite or a redox indicator and/or its cofactor discussed above, that is observed by a measuring instrument such as a spectrometer. As also noted above, redox balance is due to relative oxidation/reduction status between redox couples operating at the physical chemistry level. Some of the suitable couples without their co-factors are listed in Tables 3, 4 and 5 below. More precisely, Tables 3-5 provide an exemplary and partial list of analytes that are measured in the biomarker measurements of the instant teachings.

TABLE-US-00003 TABLE 3 Redox Pairs Analytes Panel 1 Cystine* Cysteine* Cysteine Persulfide* GSSG* GSH* GSH Persulfide* HomoCystine* XOMA H.sub.2S* Thiosulfate* Tetrathionate CysGly Dipeptide* GluCys Dipeptide* Cys-GSH Disulfide Ophthalmic Acid* Cystathionine Lanthionine GSH-Sulfonic Acid Lipoic Acid Cysteamine Methionine* Adenosine* SAM* SAH Spermine* Spermidine* Citrulline* Ornithine Kynurenine Kynurenic Acid Serine Taurine* Pyroglutamic Acid -Aminobutyric Acid* 3-NitroTyrosine* 3-ChloroTyrosine* Glutamate Homocitrilline Aspartate *Isotopically Labeled Standard used

TABLE-US-00004 TABLE 4 Redox Pairs Analytes Panel 2 NAD NADP AMP ADP ATP CAMP Xanthine Hypoxanthine* 2-deoxy-guanosine* Inosine Acetyl-Carnitine* Carnitine NADH NADPH Urate 8-OH-dG Pyrimido purinone Fumurate* Succinate* Lactate* Pyruvate* Acetoacetate 3-Hydroxybutyric Acid 743-OH 743* 886 A0001-OH A0001* -TOC -CEHC -CEHC 743-OH-Sulfate 743-OH-Gluc A0001-OH-Sulfate A0001-OH-Gluc 589* 589-OH 589-Sulfate 589-Gluc *Isotopically Labeled Standard used

TABLE-US-00005 TABLE 5 Redox Pairs Analytes Panel 3 CoQ10 Ubiquinol (CoQ10-OH) Docosahexaenoic Acid (DHA)* Arachidonic Acid (AA)* Linoleic Acid Palmitoyl Carnitine Prostaglandin E2* tetranor PGE-M* tetranor PGA-M 15-Deoxy-PGJ2 15-Deoxy-PGJ2-GSH Leukotriene E4* Leukotriene C4 8-iso-PGF2a* Creatinine (urine) 2,3-DPG (RBC contamination of plasma) *Isotopically Labeled Standard used

[0110] Particularly useful analytes measured by sensor mechanisms 108 include the presence or concentration of an oxidoreductase or of an oxidoreductase co-factor. Others include the presence or concentration of balance influencer compounds, electron balance influencer compositions or still other redox-active compounds. Tables 3-5 above provide only a partial list of such analytes that are measured by sensor mechanisms 108 of FIG. 2. A full list of such measurable analytes includes tens of thousands or more compounds and will be accessible to a person of average skill. Still other analytes of interest for measurements by sensors 108 of FIG. 2 include pK values, pH values, threshold values, context measures and soft indicators.

[0111] In the preferred embodiment, after performing repeated reference runs or experiments per above explanation, reference models 140 are learned by learning module 120 by processing time series data of reference biomarker measurements 110A-Z. These measurements are collected at times t.sub.1, t.sub.2, . . . , t.sub.i from reference in vitro samples 103A-Z. More specifically, processing module 120 learns a reference model 140A1 that is able to predict a labeled set of mitochondrial dysfunction or dysfunctions, such as D1 in Table 6 above, present in in vitro samples 103A-Z.

[0112] The set of mitochondrial dysfunction/dysfunctions are deliberately induced into in vitro cultures 102A-Z as a result of the introduction of the specific mitochondrial inhibitor, inducer, stress/stressor or insult that is known for causing the respective mitochondrial dysfunction/dysfunctions. Recall from above, that the mitochondrial inhibitor is preferably introduced in varying dosages. Model 140A1 may be learned or trained using a specific supervised machine learning algorithm such as linear regression, logistic regression, support vector machines (SVM), decision trees, random forests, neural networks and the like. Trained model 140A1 is able to predict whether the labeled mitochondrial dysfunction, for example D1, that the model was learned/trained to predict, is present or not in an unseen biomarker measurement processed in the future.

[0113] In practice, a number of additional reference models 140A2, 140A3, . . . may be learned each predicting the same labeled set D1 of mitochondrial dysfunction/dysfunctions, and each using a different machine learning algorithm. For example, reference model 140A1 may use linear regression to predict D1, model 140A2 may use logistic regressions to predict D1, model 140A3 may use Support Vector Machines (SVM) to predict D1, model 140A4 may use random forests to predict D1 and so on.

[0114] Then, learning module 120 may use the output of this ensemble of models to make a determination as to whether a biological sample from which a biomarker measurement is obtained, has mitochondrial dysfunction/dysfunctions D1 or not. For this purpose, learner 120 may choose the prediction that is most agreed upon by the ensemble of models 140A1, 140A2, 104A3, 140A4 above to predict D1. Specifically, if the majority of the models predict that the sample contains D1, then the sample is presumed to contain mitochondrial dysfunction(s) D1, otherwise not.

[0115] Similarly, by performing another set of runs or experiments with another mitochondrial inhibitor, processing module 120 deploys the above or another combination of machine learning algorithms to learn reference models, for example, models 140A5-A10 that are able to predict another labeled set D2 of mitochondrial dysfunction/dysfunctions in in vitro cultures 102A-Z. Again, D2 is present in the cultures because of the introduction of the respective mitochondrial inhibitor that is known for causing dysfunction(s) D2. Recall from above, that the mitochondrial inhibitor is preferably introduced in varying dosages.

[0116] Trained reference models 140A5-A10, each using a different machine learning algorithm, can predict the presence/absence of D2 in a given biological sample. As before, learner 120 determines the sample to possess D2 if the majority from the ensemble of models 140A5-A10 agree on the presence of D2, otherwise not. Alternatively, learner 120 may use some other suitable metric for voting on the models other than the majority. For example, prediction from certain models in the ensemble may be preferred over others by using weights. As such, an overall weighted and normalized probability from the ensemble for predicting a dysfunction D2 may be produced using the following equation:

[00001]$.Math.\mathrm{Eq}..Math.1$$\underset{1}{\overset{\mathrm{No}..Math.\mathrm{of}.Math..Math.\mathrm{models}.Math..Math.\mathrm{predicting}.Math..Math.D.Math..Math.2}{.Math.}}.Math.\frac{\mathrm{Weight}.Math..Math.\mathrm{of}.Math..Math.\mathrm{model}.Math..Math.\mathrm{predicting}.Math..Math.D.Math..Math.2*\mathrm{predicted}.Math..Math.\mathrm{prob}..Math.\mathrm{of}.Math..Math.D.Math..Math.2}{\mathrm{No}..Math.\mathrm{of}.Math..Math.\mathrm{models}.Math..Math.\mathrm{predicting}.Math..Math.D.Math..Math.2}$

[0117] In a similar fashion, as many reference models 140A11-1N are learned as desired to predict respective labeled mitochondrial dysfunction/dysfunctions in biological samples.

[0118] In some cases, times t.sub.1, t.sub.2, . . . , t.sub.i at which reference biomarker measurements are taken are selected to mark distinct stages, transitions, reaction periods or still other important times in the bioprocess of interest that the reference in vitro culture is undergoing. These include the lag, growth, stationary and death phases of the cell-culture. FIG. 3 shows the typical lag, growth or exponential, stationary and death phases of a cell-culture. Times t.sub.1, t.sub.2, . . . , t.sub.i at which the instant reference samples are drawn for measurements may thus be taken to coincide with the beginning and/or end of one or more stages. Alternatively, or in addition, they may coincide with any number of time instants during the one or more stages shown in FIG. 3 of the cell-culture. As already noted, typically the mitochondrial inhibitor is introduced at the start of the stationary phase of the cell-culture. As such, times t.sub.1, t.sub.2, . . . , t.sub.i at which the instant reference samples are drawn for measurements will be times since the start of the stationary phase of the culture.

[0119] The biomarker measurements may be made on short time scales in comparison to Gene-Protein-Reaction (GPR) times although this is not a requirement. Hence in advantageous embodiments times t.sub.1, t.sub.2, . . . , t.sub.i at which reference biomarker measurements are taken at a frequency of at least once every month, at least once every two weeks, at least once every 10 days, at least once every 5 days, at least once every 2 days, at least once every day, at least once every 12 hours, at least once every hour, at least once every 30 minutes, at least once every 10 minutes, at least once every 5 minutes, at least once every minute, at least once every 30 seconds, at least once every 10 seconds, at least once every 5 seconds, at least once every second, at least twice every second, at least 5 times every second, at least 10 times every second, at least 20 times every second, at least 50 times every second, at least 100 times every second, or more.

[0120] As already mentioned, diagnostic platform 100 of FIG. 2 can employ one or more learning methods. Some particularly useful methods in the embodiments of the present invention include Artificial Intelligence (AI) methods, Hidden Markov methods and Deep Learning (multi-layered neural network) methods.

[0121] Let us now look at the machine learning process employed by system 100 in greater detail. Specifically, the learning process is embodied by learning phase 52 of protocol 50 of FIG. 1 discussed above. The below explanation is provided rigorously for some of the machine learning techniques. Based on the framework provided below, one skilled in the art will be able conceive additional techniques of machine learning to practice the instant teachings.

Learning

[0122] Recall that processing module 120 learns a reference model, for example reference model 140A1 of FIG. 2, by deploying one or more machine learning algorithms. The reference model is capable of predicting a given labeled set of mitochondrial dysfunction/dysfunctions such as D1 induced in reference in vitro cultures 102 as a result of the varying-dosage based introduction of a specific mitochondrial inhibitor.

[0123] One way to state the objective of the learning process is for reference model 140A1 to extract the most conserved or representative biomarker fingerprint or signature across the most number of reference in vitro samples 103 of different cell-cultures 102 from different cell-lines. If such a fingerprint is also not present in the uninhibited reference in vitro culture and its predicted value is above a predetermined threshold, then it is indicative of the presence of the corresponding labeled mitochondrial dysfunction(s) D1. Stated differently, reference model 140A1 is trained to make a prediction of the presence of dysfunction D1 based on the most conserved/representative biomarker fingerprint/signature or pattern across a variety biomarker measurements 110A-Z made on samples 103A-Z drawn from a variety of cultures 102A-Z at a variety of times t.sub.i under a variety of conditions such as varying dosages di of the mitochondrial inhibitor used.

[0124] The reference biomarker measurements consist of the quantities of analytes measured, examples of which were given in Tables 3-5 above, and performed by a measuring instrument(s) 108 such as a high-throughput mass spectrometer. FIG. 4 provides an exemplary read-out or mass spectrum of measurements of some of the analytes whose measured quantities we will only denote by q1, q2, . . . instead of using their full medical names as provided in Tables 3-5. This is to avoid detraction from the main principles being taught. X-axis of FIG. 4 represents the familiar mass-to-charge ratio (m/z) and y-axis represents the relative quantities of the analytes. The measurements of some of the analytes are marked in FIG. 4 as shown. A readout from a sample, such as that shown in FIG. 4 constitutes a biomarker measurement 110 of FIG. 2 according to the present teachings.

[0125] Linear Regression:

[0126] As stated, one of the algorithms that may be used to train or learn reference model, such as model 140A1 of FIG. 2, is linear regression with supervised learning. Those skilled in the art will appreciate that linear regression is given by the equation:

Y=f(X)+Eq. 2

[0127] Here X specifies the input or independent variables, Y is the output or dependent variables or responses, f describes the relationship between X and Y and e is the random error term (positive or negative) with a mean of zero.

[0128] According to the present technology, each biomarker measurement 110 such as that shown in FIG. 4, constitutes an observation xX. X is a multi-dimensional matrix or tensor of the form [(t1, t2, . . . ti), (d1, d2, . . . dj), (c1, c2, . . . cl)], where each entry of X is a vector of the form [q1, q2, . . . qk]. Here, t1, t2, . . . ti are the times since exposure of cultures 102A-Z to a specific mitochondrial inhibitor or perturbation at which respective samples 103A-Z were drawn per above discussion. d1, d2, . . . dj are the dosage amounts in which the mitochondrial inhibitor was introduced and q1, q2, . . . qk are the quantities of specific analytes that were measured from samples 103A-Z and examples of which were given in Tables 3-5 (see also FIG. 4 and related explanation).

[0129] Further, variable c.sub.1 . . . l (where l is the small-case alphabetical letter l as in lima) is used to denote the number of cell-lines from which cultures 102A-Z for a given set of runs or experiments are grown. Recall that the cultures are grown from different cell-lines and are exposed to different dosages of the inhibitor/stressor. Then samples are taken from the cultures at different times and biomarker measurements 110 are taken from those samples. The resulting 4.sup.th order tensor X is visualized in FIG. 5 in a pictorial form by piling up three-dimensional hyperrectangles or parallelepipeds composed of rectangles as shown. The reader is cautioned not to judge the pictorial illustration of FIG. 5 which is provide for explanatory purposes too strictly. This is because the 4.sup.th order tensor X is inherently multi-dimensional and is hard to illustrate in the two-dimensional drawing of FIG. 5. As such, FIG. 5 should be taken as an intuitively convenient rather than a mathematically strict representation of tensor X.

[0130] The hyperrectangles shown in FIG. 5 are built from piling up two-dimensional matrices shown as rectangles whose rows indicate times of measurements t1-i and whose columns indicate the relative quantities q1-k of individual analytes being measured as shown in the exemplary illustration of FIG. 4. Note that each column of the two-dimensional matrices constitutes a biomarker measurement 110A-Z of the present teachings and is a vector x of measured relative quantities of the form [q1, q2, . . . qk] introduced above and shown in FIG. 4. Also, from FIG. 4 we know that these quantities are relative quantities of the analytes from the overall sample as measured by a preferably high-throughput and high-resolution mass spectrometer.

[0131] Each matrix or rectangle in FIG. 5 is populated for a given dosage quantity from dosages d1-j of the mitochondrial inhibitor being used and while using a given cell-line from cell-lines c1-l from which the in vitro cultures were grown for the set of experiments. Piling up individual matrices or rectangles gives us a three-dimensional rectangle or a hyperrectangle or a parallelepiped. Thus, each hyperrectangle is populated with vectors [q1, q2, . . . , qk] for varying times t1-i, for varying dosages d1-j but for a given cell-line from cell-lines c1-l from whom the in vitro cultures used in the set of experiments were grown. Finally, piling up individual hyperrectangles as shown by the upward facing vertical dotted arrow 180 gives us the multi-dimensional data structure constructed for our 4.sup.th order input tensor X. Thus, tensor X has i*j*l entries of vectors x each of the form [q1, q2, . . . qk] and thus having k entries each.

[0132] Based on this framework of multi-dimensional data, one skilled in the art will be able to conceive extensions of the design to incorporate additional dimensions. More specifically, one can conceive extending our tensor X of FIG. 5 with additional dimensions that represent additional varying conditions of the experiments/runs. For example, one can have another dimension tempi to represent varying temperatures that cultures 102 were subjected to at which samples 103 were drawn to make measurements 110. Similarly, one can have a dimension pi to represent the pressure that the samples were subjected to at which the measurements were taken, another dimension gi to represent exposure to a gas specie, etc. However, to not detract from the key principles being taught, the below teachings will be based on the 4.sup.th order tensor X shown in FIG. 5 and discussed above.

[0133] Y in Eq. (2) above is the labeled mitochondrial dysfunction or the labeled set of mitochondrial dysfunctions being predicted by learning system 100 of FIG. 2. System 100 is trained on labeled data constructed for tensor X as shown in FIG. 5. Examples of mitochondrial dysfunctions and their associated labels Dm were given in Table 2 above. These dysfunctions are induced as a result of the introduction as perturbations or stresses of respective mitochondrial inhibitors specified in the column named Mitochondrial dysfunction induced of Table 2 above.

[0134] Though there may be more than one dysfunctions induced by a given inhibitor as shown in Table 2 above (for example D5), we will collectively refer to all the dysfunctions associated with a given mitochondrial inhibitor by a singular label Dm and at times refer to these dysfunctions only in the singular, as dysfunction. It is these labeled dysfunction(s), such as D1, that model 140A1 predicts as dependent variables or responses Y in Eq. (2) above.

[0135] Since the present embodiment uses linear regression, Y is assumed to be linearly related to X. Further, since X has more than one values or features, the regression technique is termed as multiple linear regression, and for our tensor X above, it is given by:

Y=.sub.0+.sub.1x.sub.(t1,d1,c1)+.sub.2x.sub.(t2,d1,c2)+.sub.i*j*lx.sub.(ti,dj,cl)Eq. 3A

[0136] Here, x(t.sub.1 . . . i, d.sub.1 . . . j, c.sub.1 . . . l)X are the input variables or features illustrated in FIG. 5 and are each vectors of the form [q1, q2, . . . qk] noted above. In the present embodiment, these are the measured quantities of the various analytes q1-k (see Table 3-5 above) constituting a biomarker measurement taken from a sample drawn at a given time ti from an in vitro culture grown from a cell-line cl (see Table 1 above) exposed to a given dosage dj of a specific/chosen mitochondrial inhibitor (see Table 2 above). Each measurement taken is a collection of read-outs of the various quantities of analytes qk such as the ones exemplarily given in Table 3-5 above and illustrated in the exemplary mass spectrum or readout of FIG. 4.

[0137] In Eq. (3A), .sub.0, .sub.1, . . . , .sub.i*j*l are the coefficients of the linear regression to be determined as further explained below. An error term .sub.1 . . . i*j*l, is also added to Eq. (2) above and it indicates the quality of the prediction of Y by the model as compared to the actual value of Y as measured. This error term can be estimated as root-mean-squared error (RMSE) between the predicted/hypothesized and actual values of Y.

[0138] In order to train a given reference model, we shall first divide the known dataset represented by tensor X of the above discussion, into two parts. The first part preferably constitutes 80% of the total entries in X and is referred to as training data/dataset.

[0139] The second part preferably constitutes the remaining 20% entries in the dataset and is referred to as test data/dataset. In this manner, reference models 140A1-AN are trained or learned on the training dataset and are then tested on the test dataset to determine their accuracy and bias. As will be further discussed, this allows one to tune the hyperparameters of the models and address their bias-variance tradeoff with techniques known in the art.

[0140] Note that since Y in Eq. (3A) represents the response or predicted value of a single labeled mitochondrial dysfunction(s) for a given set of experiments, it is as such a vector of probabilities. Reference models 140A1-AN are each trained by providing it a value from D1, D2, . . . , Dm after the measurement of each biomarker measurement of relative quantities represented by vector x=[q1, q2, . . . qk]. In other words, in the exemplary case of training our model 140A1 to predict D1, one would provide model 140A1 the known label D1 in response to each observation x=[q1, q2, . . . qk]X corresponding to each of i*j*l vector entries x of tensor X. As such, vector Y also has i*j*l entries and in its expanded form as denoted by:

yY[y.sub.x(t1,d1,c1), . . . ,y.sub.x(ti,d1,c1),y.sub.x(t1,d2,c1), . . . ,y.sub.x(ti,d2,c1), . . . ,y.sub.x(t1,dj,c1), . . . ,y.sub.x(t2,dj,c1), . . . ,y.sub.x(ti,dj,c1), . . . ,y.sub.x(t1,dj,c2), . . . ,y.sub.x(ti,dj,cl)],

[0141] or more simply by Y.sub.t1 . . . i,d1 . . . j,c1 . . . l or still as a vector Y of the form [y1, y2, . . . , y.sub.i*j*l]. Thus, a partial excerpt from an exemplary training data derived from input tensor X for system 100 may be given by Table 7 below:

TABLE-US-00006 TABLE 7 Training dataset for dysfunction D1 with inhibitor A8674 No. Y X (t.sub.i) X (d.sub.j) x (c.sub.l) X (q.sub.k) 1 0% 1 minute 0 Y79 2 0% 5 minutes 0 Y79 3 0% 30 minutes 0 Y79 4 0% 5 hours 0 Y79 5 0% 1 day 0 Y79 6 0% 1 minute 1 uL Y79 7 50% or 0.5 5 minutes 1 uL Y79 8 100% or 1 30 minutes 1 uL Y79 9 100% or 1 1 hours 1 uL Y79 10 100% or 1 5 hours 1 uL Y79 11 100% or 1 10 hours 1 uL Y79 12 100% or 1 2 days 1 uL Y79 13 50% or 0.5 1 minute 5 uL Y79 14 100% or 1 5 minutes 5 uL Y79 15 100% or 1 30 minutes 5 uL Y79 . . . . . . . . . . . . . . . . . .

[0142] In the above training dataset, each entry of column Y is a value in the vector Y discussed above. Each entry {circumflex over (q)} of column X(qk) with a hat {circumflex over ()} denotes a vector and is an entry in input tensor X, and is a vector of measured quantities [q1, q2, . . . , qk] of the analytes in the biomarker measurement such as those shown in Tables 3-5. In other words, the training dataset consists of measured vectors {circumflex over (q)} at indices i,j,l of tensor X per above explanation. Human Caucasian retinoblastoma (Y79) from Table 1 above is used as the cell-line from which in vitro cultures are grown. Column X(d) indicates dosage of the introduction of the mitochondrial inhibitor, for example, A8674 (Antimycin A from Streptomyces sp.) from Table 2 above and Column X(t) indicates the times at which the samples were drawn for measurements.

[0143] Note that rows 1-6 correspond to the scenario when no inhibitor was exposed (zero/O state), and unsurprisingly column Y indicates a 0 or a 0% probability that dysfunction D1 is present in the corresponding culture. Row 7 indicates that at 1 minute exposure of a dosage of 1 uL of A8674, the culture is known not to have developed dysfunction D1 or still has a 0% chance, at 5 minute exposure it has a 50% chance and at 30 minutes or above, it has a 100% chance of developing dysfunction D1. However, with a dosage of 5 uL of A8674, the culture has a 50% chance of D1, at 5 minutes or above, it has a 100% of chance of having D1. Recall that it is important to have rows 1-6 in the dataset with no introduction of the inhibitor, so model 140A1 can learn what patterns or biomarker fingerprints do not correspond to D1.

[0144] Now let us understand how diagnostic platform determines the values of weights or coefficients .sub.0, .sub.1, . . . , .sub.i*j*l in Eq. (3A) above. Determining these coefficients or internal parameters is necessary in order to compute using Eq. (3A) the value of Y indicative of the presence or predictive of a specific labeled dysfunction(s) D1-m per above explanation. This allows for the training or learning of model 140A1 of the present example on a given training dataset/data. Once model 140A1 has been trained, or alternatively stated, the values of coefficients/weights .sub.0, .sub.1, . . . , .sub.i*j*l have stabilized or converged as will be taught below, Eq. (3A) is then used to predict the presence of D1 on unseen biological samples in the future.

[0145] Depending on the values of these coefficients/weights .sub.0, .sub.1, . . . , .sub.i*j*l, model 140A1 is able to estimate which observations x=[q1, q2, . . . qk]X corresponding to each of i*j*l vector entries x of tensor X are more important than others. Because of varying scales/units of features/observation x, it is generally ill-advised to ascribe the importance of an observation x on the basis of the value of its estimated coefficient .sub.x{.sub.0, .sub.1, . . . , .sub.i*j*l}computed using Eq. (3A) above. However, one can use standardized regression coefficients for comparison.

[0146] Those skilled in the art will appreciate that standardized regression coefficients are the estimates resulting from a regression analysis that have been standardized so that the variances of dependent and independent variables are 1. More rigorously, they mean how many standard deviations a dependent variable will change, per standard deviation increase in the predictor variable. A standardized coefficient * is derived from estimated coefficient .sub.x by using the formula:

*.sub.x=.sub.x.Math..sub.x/.sub.yEq. 3B

[0147] In other words, multiply the estimated coefficient from Eq. (3A) by the standard deviation of the predictor variable xX and divide it by the standard deviation of the outcome variable yY to arrive at the standardized coefficient. Then, values of standardized coefficients * that are above a predetermined threshold signify the biomarker fingerprint/signature that is indicative of mitochondrial dysfunction D1.

[0148] Furthermore, these selected standardized coefficients * may then be used to reduce/select the observations and consequently those analytes that are more predictive of dysfunction D1 than others. This feature selection/reduction allows one to choose more specialized or targeted sensing equipment during targeting phase 54 of protocol 50 (see FIG. 1). Also, this leads to economies of scale or practice by not tying up expensive equipment, such as a high throughput and high-resolution spectrometer, by requiring it to only measure those analyte peaks that matter most for prediction of specific mitochondrial dysfunctions.

[0149] Stated differently, as learning module/engine 120 trains models to predict labeled mitochondrial dysfunctions, uncorrelated or noisy measurements are dropped out. The models can thus detect patterns of analytes or fingerprints that best predict the labeled dysfunctions based on the most relevant training data obtained during learning phase 52 of FIG. 1 in vitro. These models can then be applied to a reduced set of measurements that may be available in vivo during targeting phase 54 and without compromising the predicting prowess of the models.

[0150] Ordinary Least Squares:

[0151] In one embodiment, system 100 learns reference models such as model 140A1 by employing ordinary least squares (OLS) for solving Eq. (3A) above. This approach can lead to a completely analytical and closed-form solution. More rigorously, Eq. (3A) is written with a residuals term e for some estimate of model parameters 3 as follow:

Y=.sub.0+.sub.1x.sub.(t1,d1,c1)+.sub.2x.sub.(t2,d1,c2)+.sub.i*j*lx.sub.(ti,dj,cl)+e.sub.i*j*l.Math.e.sub.i*j*l=Y(.sub.0+.sub.1x.sub.(t1,d1,c1)+.sub.2x.sub.(t2,d1,c2)+.sub.i*j*lx.sub.(ti,dj,cl))Eq. 4

[0152] Or more simply, by dropping the subscripts and including a hat {circumflex over ()} over coefficients/parameters vector 3 to indicate a vector, we get:

e=YX{circumflex over ()}Eq. 5

[0153] The above assumes that the first value x(0,0,0) in tensor X is a unit vector of the form [1,0,0 . . . ]. Such an assumption leads to .sub.0 being the intercept of the regression in the hyperspace of our tensor X. It follows from Eq. (5) above that:

[00002]$\begin{array}{c}{e}^T.Math.e=.Math.{\left(Y-X.Math..Math.\hat{}\right)}^T.Math.\left(Y-X.Math.\hat{}\right)\\ =.Math.Y^T.Math.Y-Y^T\left(X.Math..Math.\hat{}\right)-{\left(X.Math.\widehat{.Math.}\right)}^T.Math.Y+{\left(X.Math..Math.\hat{}\right)}^T.Math.\left(X.Math.\hat{}\right)\\ =.Math.Y^T.Math.Y-{\left(X.Math..Math.\hat{}\right)}^T.Math.Y-{\left(X.Math.\hat{}\right)}^T.Math.Y+{\left(X.Math..Math.\hat{}\right)}^T.Math.\left(X.Math.\hat{}\right)\\ =.Math.Y^T.Math.Y-2.Math.{\left(X.Math..Math.\hat{}\right)}^T.Math.Y+{\left(X.Math..Math.\hat{}\right)}^T.Math.\left(X.Math.\hat{}\right)\\ =.Math.Y^T.Math.Y-2.Math..Math.\hat{}.Math.X^T.Math.Y+.Math.{\hat{}}^T.Math.X^T.Math.X.Math.\widehat{.Math.}\end{array}$

[0154] Here superscript T represents a transpose operation and for our tensor X, it is obtained by preserving the first index i and by swapping subsequent indices j and l while also preserving each vector entry xX in its original form. In other words, each vector xX is preserved as is, the vectors at indices i are also retained as they are, and vectors at each value of indices j and l are swapped. For a tutorial on higher order tensor operations, the reader is referred to Higher Order Tensor Operations And Their Applications by Emily Miller of The College of New Jersey and Scott Ladenheim of Syracuse University, published in TCNJ Journal Of Student Scholarship Volume XI, dated 2009.

[0155] To determine the coefficients {circumflex over ()}, we minimize the sum of squared residuals with respect to the parameters:

[00003]$\begin{array}{c}\frac{\left[e^T.Math.e\right]}{}=.Math.0\\ =.Math.-2.Math.X^T.Math.Y+2.Math.X^T.Math.X.Math..Math.\hat{}\end{array}$

[0156] Utilizing the identity

[00004]$\frac{a^T.Math.b}{a}=b$

for vectors a and b, we get:

X.sup.TY=X.sup.TX{circumflex over ()}Eq. 6

The above directly leads to:

{circumflex over ()}=(X.sup.TX).sup.1X.sup.TY,Eq. 7

[0157] provided (X.sup.TX) is invertible or non-singular. Eq. (6) and (7) above are in the form of Normal Equations for coefficients {circumflex over ()} and can be solved analytically since the values of X and Y are known. Thus, in the present embodiment, model 140A1 is trained by using tensor X of measurements/observations of biomarker measurements in Eq. (7) above to compute the values of coefficients or parameters or weights f of the linear regression.

[0158] Then, given a new and unseen value of tensor X containing test data, model 140A1 is able to predict the value of a mitochondrial dysfunction/dysfunctions D1 (Table 2 above) by using Eq. (3A) above. Recall that we had partitioned input tensor X into a training dataset and a test dataset. Thus, we can train model 140A1 on the training dataset and compute its bias and accuracy on the test dataset and thusly tune its hyperparameters, as will be further discussed below.

[0159] Since the training data contains values of Y that are normalized as probabilities between 0 and 1 (see Table 7 above), Eq. (3A) will produce values of Y that are also normalized between 0 and 1. As such, these values are interpreted to mean the probability of the presence or absence of dysfunction D1 in the test data. Thus, in the present embodiment, we are using linear regression for classification.

[0160] In other words, if a given value yY=[y1, y2, . . . y.sub.i*j*l] is 0.7 then there is a 70% chance that the corresponding measurement in input dataset X is indicative of dysfunction D1 in corresponding cell-culture from which measurement sample was obtained. Recall that input tensor X contains i*j*l entries of vectors x, each of the form [q1, q2, . . . , qk]. Furthermore, the test dataset derived from tensor X may contain one or more vectors x=[q1, q2, . . . qk]X each producing a corresponding value yY. Although as noted above, it is advisable to partition input tensor X such that test dataset is about 20% of the size of the overall input data X.

[0161] Gradient Descent:

[0162] If the number of features or in our case vectors in input tensor X is very large, for example, greater than 10,000, then Eq. (7) can have a high computation time. Therefore, in another embodiment, the technique of gradient descent is employed to compute coefficients {circumflex over ()}.

[0163] In this embodiment, a cost function is iteratively minimized to arrive at the estimated values of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l. More rigorously, let us denote by H.sub. the predicted or hypothesized value of a given dysfunction, for example dysfunction D2 (Table 2 above), by a reference model, such as model 140A5. The purpose of doing this is so that we can differentiate the predicted or hypothesized value h.sub.H.sub. based on coefficients {circumflex over ()} from the corresponding actual value yY as measured or known in the dataset.

[0164] In other words, we will denote the left-hand side of Eq. (3A) by H.sub. and rewrite it as:

H.sub.=X{circumflex over ()}Eq. 8

[0165] It should be apparent that each of vectors H.sub. and {circumflex over ()} have i*j*l entries. In other words, h.sub.1 . . . i*j*lH and y.sub.1 . . . i*j*lY. We will now define the cost or loss function J() that associates a cumulative cost to the difference between the predicted value h.sub.1 . . . i*j*lH obtained by Eq. (8) and the corresponding actual value y.sub.1 . . . *j*lY as known. Specifically,

[00005]$\begin{array}{cc}J\left(\right)=\frac{1}{i*j*l}.Math.\underset{a=1}{\overset{i*j*l}{.Math.}}.Math.{\left(h_{}.Math.x_a-y_a\right)}^2& \mathrm{Eq}..Math.9\end{array}$

[0166] By minimizing J() in successive iterations, we can obtain estimates of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l to be used in Eq. (3A) above on test or unseen data in the future. Specifically, we perform gradient descent to find the local minimum of J() using the equation:

[00006]$\begin{array}{cc}{}_{1.Math..Math.\mathrm{.Math.}.Math..Math.i*j*l}={}_{1.Math..Math.\mathrm{.Math.}.Math..Math.i*j*l}-.Math.\frac{}{{}_{1.Math..Math.\mathrm{.Math.}.Math..Math.i*j*l}}.Math.J\left(\right)& \mathrm{Eq}..Math.10\end{array}$

[0167] Here is the learning rate hyperparameter indicative of how aggressively the process tries to find the minimum or convergence. It can be chosen very conservatively at the outset and may be gradually increased later to achieve convergence quicker. An exemplary initial value of a may be taken as 0.000001. Care should be taken not to increase a too sharply or the algorithm may overshoot the minimum and diverge.

[0168] According to the present design, the gradient descent process of estimating values of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l is given by the following pseudocode: [0169] Step 1: Initialize all coefficients .sub.0, .sub.1, . . . , .sub.i*j*l to an initial value, for example 0. [0170] Step 2: Compute predictions H.sub. by Eq. (8) [0171] Step 3: Compute loss J() by Eq. (9) [0172] Step 4: Update coefficients .sub.0, .sub.1, . . . , .sub.i*j*l by Eq. (10) [0173] Step 5: Are the values of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l similar to their values from the previous iteration within a predetermined threshold, for example, 0.1 ? If Yes, Exit. If Not, go back to Step 2.

[0174] Continuing with our above example, once the values of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l are stabilized or in other words gradient descent has converged, then our model 140A1 is said to have been trained to predict dysfunction D1, using Eq. (8) above. To address the familiar problem of over-fitting, a regularization term may be added to cost function J() of Eq. (9) as follows:

[00007]$\begin{array}{cc}J\left(\right)=\frac{1}{2*i*j*l}\left[\underset{a=1}{\overset{i*j*l}{.Math.}}.Math.{\left(h_{}.Math.x_a-y_a\right)}^2+.Math.\underset{b=1}{\overset{i*j*l}{.Math.}}.Math.{}_b^2\right]& \mathrm{Eq}..Math.11\end{array}$

[0175] Here is the regularization coefficient and is another hyperparameter that can be tuned. A higher X will more harshly penalize large coefficients that could lead to potential overfitting.

[0176] In alternative embodiments, Newton's method may be utilized to minimize loss J() and estimate the values coefficients .sub.0, .sub.1, . . . , .sub.i*j*l. Using the above-provided framework and our input tensor X, one skilled in the art is also able to apply other variations of these algorithms and related machine learning techniques, such as batch gradient descent, and the like. Furthermore, and as already noted, one can extend input tensor X to include additional dimensions that represent environmental conditions, or other variables of the experiments.

[0177] Logistic Regression:

[0178] In still other embodiments, the machine learning technique of logistic regression may be employed as a natural choice for classification of samples into those that have the labeled dysfunction and those that do not. Again, since input X has multiple features, the technique may be termed as multiple logistic regression. To use logistic regression, we define p=E(yY|xX) as the conditional probability y for some value of regressors xX. In other words, p defines the probability that a dysfunction Dm is present in a given biomarker measurement x measured from a sample. Then, logistic or logit regression given our tensor X above is expressed by the equation:

[00008]$\begin{array}{cc}\ln .Math.\left[\frac{p}{1-p}\right]={}_0+{}_1.Math.x_{\left(t.Math..Math.1,d.Math..Math.1,.Math.c.Math..Math.1\right)}+{}_2.Math.x_{\left(t.Math..Math.2,d.Math..Math.1,c.Math..Math.2\right)}+\mathrm{.Math.}+{}_{i*j*l}.Math.x_{\left(\mathrm{ti},\mathrm{dj},\mathrm{cl}\right)}& \mathrm{Eq}..Math.12\end{array}$

[0179] Once again, gradient descent is used to compute the values of coefficients or weights .sub.0, .sub.1, . . . , .sub.i*j*l. More specifically, denoting hypothesized values again by h.sub.H.sub., we can rewrite Eq. (12) above as follows:

[00009]$\begin{array}{cc}\ln .Math.\left[\frac{h_{}}{1-h_{}}\right]={}_0+{}_1.Math.x_{\left(t.Math..Math.1,d.Math..Math.1,.Math.c.Math..Math.1\right)}+{}_2.Math.x_{\left(t.Math..Math.2,d.Math..Math.1,c.Math..Math.2\right)}+\mathrm{.Math.}+{}_{i*j*l}.Math.x_{\left(\mathrm{ti},\mathrm{dj},\mathrm{cl}\right)}& \mathrm{Eq}..Math.13\end{array}$

[0180] We can then define the regularized cost function as:

[00010]$\begin{array}{cc}j\left(\right)=-\frac{1}{i*j*l}.Math.\underset{a=1}{\overset{i*j*l}{.Math.}}.Math.\left[y_a.Math.\ln \left[h_{}.Math.x_a\right]+\left(1-y_a\right).Math.\ln \left[1-h_{}.Math.x_a\right]\right]+\frac{}{2*i*j*l}.Math.\underset{b=1}{\overset{i*j*l}{.Math.}}.Math.{}_b^2& \mathrm{Eq}..Math.14\end{array}$

[0181] As with linear regression, here is the regularization coefficient and is a hyperparameter that can be tuned to control how harshly overfitting is penalized. We can then use the following pseudocode to compute the values of coefficients .sub.0, .sub.1, . . . .sub.i*j*l. [0182] Step 1: Initialize all coefficients .sub.0, .sub.1, . . . , .sub.i*j*l to an initial value, for example 0. [0183] Step 2: Compute predictions H.sub. by Eq. (13) [0184] Step 3: Compute loss J() by Eq. (14) [0185] Step 4: Update coefficients .sub.0, .sub.1, . . . , .sub.i*j*l by Eq. (10) [0186] Step 5: Are the values of coefficients .sub.0, .sub.1, . . . , .sub.i*j*l similar to their values from the previous iteration within a predetermined threshold, for example, 0.1 ? If Yes, Exit. If Not, go back to Step 2.

[0187] In alternative embodiments, Newton's method may be used to minimize loss J() and to estimate the values coefficients .sub.0, .sub.1, . . . , .sub.i*j*l. After computing the values of .sub.0, .sub.1, . . . , .sub.i*j*l, one can now easily compute the value of p by using Eq. (12) above. One then defines a threshold value, for example 0.5 or 0.7. If the value of p is greater than the threshold value then the mitochondrial dysfunction in question, D1 from our prior example, is assumed to exist in the sample, otherwise not. In other words, the output is a labeled dysfunction Dm if p>threshold value, otherwise not. The library of reference models 140A1-AN trained according to above teachings is stored in database 170 as shown in FIG. 2.

[0188] Using the framework provided above, one skilled in the art can also apply the techniques of support vector machines (SVM) and/or decision trees to classify samples into those that have the labeled dysfunction and those that do not. With the above-provided tensor X and the related teachings of the learning framework, one can conceive the implementation details of SVM and decision trees-based learning models to practice the instant principles. As such, the SVM and decision trees implementations are not discussed in detail in this disclosure. Furthermore, one will also be able to compute the R.sup.2 and RMSE metrics for the models to determine their efficacy on test data.

[0189] One is also able to apply the familiar technique of cross-validation to more effectively fine-tune the models and their hyperparameters than just by statically partitioning input tensor X as described above. In other words, input tensor X may be partitioned into N folds or parts with training done on the first N1 folds followed by testing on the N.sup.th (held-out) fold. This is followed by training on the 2.sup.nd to N.sup.th folds combined followed by testing on the 1.sup.st fold, and then training on 3.sup.rd to N.sup.th and the 1.sup.st folds combined followed by testing on the 2.sup.nd fold, and so on. Cross-validation provides for a superior error-estimation, removal of bias and reduction of variance of the models.

[0190] Additionally, the techniques of deep learning may also be applied to learn models 140A1-AN of FIG. 2. In particular, Deep Neural Networks (DNN) may also be implemented to train the models. By utilizing an ensemble of DNN's with 3 or more hidden layers, models 140A1-AN are able to classify an input biomarker measurement as containing or not, a known mitochondrial dysfunction(s) Dm. Examples of dysfunctions Dm were given in Table 2 above. The reader is referred to the reference entitled, Deep biomarkers of human aging: Application of deep neural networks to biomarker development by Putin et al. published in AGING, Vol. 8 No. 5, dated May, 2016 for a framework for using DNN's for classification problems in aging research.

[0191] According to the present design, the biomarker fingerprint or pattern of relevant analytes in response to the inhibition of one or more mitochondrial functions by a mitochondrial inhibitor, may be conserved/represented (and observed/measured) across a range of cell-cultures. The cultures may be drawn from a diverse array of cell-lines and across varying times of the growth of the cultures. Further, the biomarker fingerprint will be absent or not conserved in uninhibited or normal cell-cultures. As explained, the determination of the most conserved reference biomarker fingerprints is performed by reference models 140A1-AN.

[0192] Since each reference model is trained to predict a specific dysfunction, ultimately the output of system 100 for a given unseen dataset consisting of one or more vectors x=[q1, q2, . . . qk] is a ranked list of dysfunctions as predicted for each vector x in the dataset. The ranking of the dysfunctions can be in the order of the strength/probability of the presence of a certain dysfunctions as predicted by the reference models trained per above explanation.

[0193] For example, if for a given unseen dataset consisting of a single vector x, the ensemble of models 140A1-A4 predicts D2 with an overall weighted/normalized probability from Eq. (1) above of 85%, and the ensemble of models 140A5-12 predict D4 with an overall weighted/normalized probability from Eq. (1) above of 75%, then system 100 will produce a ranked list of dysfunctions D2 with a probability of 85%, followed by D4 with a probability of 75% for vector x, and so on. Similarly, if the unseen data has more than one vectors x, such a ranked list can be produced for each of those vectors x.

[0194] Note that in practical clinical settings, unseen dataset derived from an in vivo subject/patient as will be explained below, will likely consist of one vector x. It makes sense that if there are more than one vectors x of unseen data, that they are all derived from the same in vivo subject/patient. In such a scenario, the vectors x may be derived from the same patient at different points in time.

[0195] The above-described techniques of learning of the present teachings make frequent use of matrix manipulations and are naturally suited for deployment on Graphical Processing Unit (GPU) based architectures. These include GPU architectures available from vendors such as Nvidia, Advanced Micro Devices (AMD), ARM Holdings, Broadcomm, Intel, Qualcomm, etc., as well as cloud-based GPU virtual services such as Google Cloud Platform, Amazon Web Services, IBM Cloud and the like. Moreover, the algorithms and the mathematical framework provided above may be implemented in a number of programming environments of choice.

[0196] These include TensorFlow, Caffe, Matlab, R, Azure, Apache Singa, H20, Scikit-Learn, etc. as well as general purpose programming languages including C, C++, Java, Python, etc.

Targeting

[0197] Having described above in detail the training or learning of models 140A1-AN to predict specific mitochondrial dysfunctions Dm, let us continue our discussion of testing with in vivo samples and refer back to FIG. 2. This testing with in vivo samples is referred to as targeting in the present design and is designated by targeting phase 54 of protocol 50 of FIG. 1 presented earlier.

[0198] During targeting, models 140A1-AN that were trained on in vitro data obtained from in vitro samples 103A1-AZ are used to make predictions on target in vivo samples. Of course, the dysfunctions thus predicted by the models are the same as the ones that the models were trained on. The in vivo target samples are obtained from target patients that are known to have genetic mitochondrial diseases characterized by various mitochondrial dysfunctions. These diseases are also referred to as simply mitochondrial diseases for short. In addition, or alternatively, the patients may be known to have these genetic mitochondrial diseases diagnosed from their sequenced genomic data.

[0199] To explain targeting further, FIG. 2 also shows target/targeting in vivo subjects/patients or biological entities 152A, 152B, . . . , 152X from which respective in vivo samples 153A, 153B, . . . , 153X are drawn. Types of in vivo samples that may be obtained from target subjects/patients include blood or blood components, urine, stool samples, pleural fluid, ascites, sputum, tissue, plasma, tears, sweat, saliva, etc. As would typically be the case, only one sample 153A is being shown drawn from subject 152A, only one sample 153B from subject 152B and so on, although that is not a requirement. Analogously to the in vitro samples of the learning phase, each sample 153A-X is measured by a respective sensor or measuring instrument 158A-X that is preferably a high-throughput high-resolution mass spectrometer. Further, sensors 158A-X may be common/shared for all the in vivo samples and/or in vivo and in vitro samples as shown by dotted line 109.

[0200] Target biological entities or subjects 152A-X from whom respective in vivo samples 153A-X are drawn, are also subjected to respective target conditions 154A-X as shown. Specifically, target in vivo samples 152A-X may be live humans, plants, animals, organisms, or any other biological entities in their respective natural in vivo environments or habitats 154A-X. Targeting conditions 154 in practice are the clinical conditions in which samples 153 are drawn from the patients/subjects.

[0201] In an analogous fashion to reference measurements 110A-Z explained above, sensor systems 158A, 158B, . . . , 158X gather target biomarker measurements 160A, 160B, . . . , 160X generated from target in vivo samples 153A, 153B, . . . , 153X respectively. Each of the in vivo targets/subjects 152A-X from whom respective in vivo samples 153A-X are extracted is known to have a genetic/mitochondrial disease(s) characterized by mitochondrial dysfunction(s). Such a diagnosis for patients/subjects 152A-X may have been made on their sequenced genomic data or DNA sequencing data shown by reference numerals 156A-X respectively in FIG. 2. Sequenced genomic data or simply genomic data 156A-X may have been obtained from these subjects/patients using DNA sequencers/sequencing devices available in the art. It contains the known genetic defects that are causal of the genetic/mitochondrial diseases known to exist in these patients. However, in other cases these patients may not exhibit a genetic defect or the corresponding gene may not have expressed itself, but still the patients are known to have a mitochondrial disease based on other clinical diagnosis.

[0202] According to the instant design, target in vivo biomarker measurements 160A-X are processed during targeting by our library of reference models 140A1-AN trained above. During targeting, the reference models predict the presence of labeled dysfunctions Dm (see Table 2 above), by analyzing target biomarker measurements 160A-X originating from respective samples 153A-X of respective patients 152A-X. Let us consider that based on target biomarker fingerprints detected by models 140A1-AN in measurements 160A-X per above teachings, a dysfunction (for example D2) is predicted to be present in a statistically significant number of matching or matched target patients 152A-X (for example 500 patients). All these matched patients are known to have the same genetic mitochondrial disease (for example Complex I Deficiency).

[0203] The above knowledge provides system 100 a mapping or association 130 of dysfunction D2 to the mitochondrial disease of Complex I Deficiency along with its causal genetic defects. If D2 is implicated in Complex 1 Deficiency, then this knowledge validates our learning models in that they have accurately predicted D2 in patients that are otherwise known to have the same dysfunction/disease. This knowledge/mapping/association heretofore not existing in prior art, also exposes the correlation that exists between the causal mitochondrial inhibitor and the genetic defects/patterns observed in these patients.

[0204] In a similar manner, the targeting process is carried out for all known dysfunctions Dm against the available target population to find statistically significant number of matches to known genetic mitochondrial diseases. Thus, system 100 learns a mapping or association 130 shown in FIG. 1-2 of each mitochondrial dysfunction Dm (see Table 2 above) with the corresponding genetic mitochondrial disease and its related genetic defects/pattern. This knowledge, heretofore not available in the prior art, is very useful, because it can lead to offering new therapies for those in vivo patients as will be discussed further below.

[0205] Sometimes patients suffering from a mitochondrial disease may still not show genes expressive of their disease in their genetic data or pattern. Even in such scenarios, the above mapping is useful for associating the predicted mitochondrial dysfunction to the genetic pattern that indeed is expressed by those patients. This approach detects fingerprints of mitochondrial inhibitors that correlate to a set/cohort of gene variations. This is a useful way of learning about gene functions.

[0206] Table 8 below provides a partial excerpt of an exemplary mapping 130 that may be generated during the targeting phase. Mapping 130 is also stored in database 170 of FIG. 2 along with the library of trained reference models 140A1-AN. In addition, genomic data 156A-X of respective target patients 152A-X may also be stored in database 170 although that is not a requirement.

TABLE-US-00007 TABLE 8 Mapping 130 No. of Matched Cause/Genetic Dysfunction Patients Genetic Disease Defect D12 876 Complex I Deficiency Autosomal D3 743 Encephalomyopathy Autosomal Recessive D10 1242 Neuropathy, Ataxia, & Mitochondrial DNA Retinitis Pigmentosa point mutations in (NARP) genes associated with Complex V: T8993G, (also T8993C by some research)

[0207] As will be described in detail in the diagnosis phase, if an exemplary patient 152C with a known genetic mitochondrial disease is predicted by system 100 to have a dysfunction D3 and if D3 has a known rescuer (for example R1), then R1 may be provided in a targeted personalized therapy for patient 152C. This is an important innovation of the present design over the prevailing art since there is a growing number of known genetic mitochondrial diseases/disorders that can benefit from the present design.

[0208] As noted above, targeting phase 54 of FIG. 1 may be carried out only on a reduced number of features or analytes that are measured based on techniques including standardized coefficients during learning phase 52. This leads to specialized measurements and economies of scale as also noted above. Furthermore, trained models 140A1-AN may be even more refined during targeting to further reduce the number of features/analytes to be measured that are most predictive of a given mitochondrial dysfunction during diagnosis phase 56 as explained further below. To accomplish this, standardized coefficients or other techniques of feature selection/reduction may again be employed. For these reasons, learning phase 52 of protocol 50 may also be referred to as an untargeted phase that leads to more focused/refined targeted/targeting phase 54 and which leads to even more focused/refined diagnosis phase 56.

[0209] Table 9 below provides a partial list of the known genetic diseases/disorders that may be benefited this way.

TABLE-US-00008 TABLE 9 No. Mitochondrial Disease Capsules: 1 Alpers Disease 2 Barth Syndrome/LIC (Lethal Infantile Cardiomyopathy) 3 Beta-oxidation Defects 4 Carnitine-Acyl-Carnitine Deficiency 5 Carnitine Deficiency 6 Creatine Deficiency Syndromes 7 Co-Enzyme Q10 Deficiency 8 Complex I Deficiency 9 Complex II Deficiency 10 Complex III Deficiency 11 Complex IV Deficiency/COX Deficiency 12 Complex V Deficiency 13 CPEO 14 CPT I Deficiency 15 CPT II Deficiency 16 KSS 17 Lactic Acidosis 18 LBSL - Leukodystrohpy 19 LCAD 20 LCHAD 21 Leigh Disease or Syndrome 22 Luft Disease 23 MAD/Glutaric Aciduria Type II 24 MCAD 25 MELAS 26 MERRF 27 MIRAS 28 Mitochondrial Cytopathy 29 Mitochondrial DNA Depletion 30 Mitochondrial Encephalopathy 31 MNGIE 32 NARP 33 Pearson Syndrome 34 Pyruvate Carboxylase Deficiency 35 Pyruvate Dehydrogenase Deficiency 36 POLG2 Mutations 37 SCAD 38 SCHAD 39 VLCAD

Diagnosis

[0210] Once models 140A1-AN of FIG. 2 have been trained and targeted per above teachings, they are ready to be deployed in the field or clinical settings as indicated by diagnosis phase 56 of protocol 50 of FIG. 1. They can be effectively used to diagnose previously undiagnosed patients of mitochondrial dysfunctions and their association with any genetic mitochondrial diseases that the patients may be possessing. This is done by analyzing a heretofore unseen biomarker measurement obtained from their clinical sample and detecting a biomarker fingerprint predictive of a known mitochondrial dysfunction and an associated/mapped genetic mitochondrial disease per above teachings. The clinical sample may consist of blood or blood components, urine, stool samples, pleural fluid, ascites, sputum, tissue, plasma, tears, sweat, saliva, etc.

[0211] More generally, once diagnostic platform 100 processes the patient's sample, models 140A1-AN produce a list of potential mitochondrial dysfunctions in the patient and any associated mitochondrial diseases per above teachings. This diagnostic process 200 using our diagnostic platform 100 is illustrated in FIG. 6.

[0212] FIG. 6 shows further architectural details of diagnostic platform 100 of FIG. 2. In particular, platform/system 100 consists of a learning module 120 that during the learning phase learns/trains a library of trained models (not shown in FIG. 6) based on in vitro reference samples 103 obtained from in vitro cultures 103 per above teachings. Then, during targeting phase, based on target samples 153 obtained from in vivo target subjects 152, targeting module 122 develops a mapping 130 of predicted mitochondrial dysfunctions to genetic diseases known to be present in the target subjects also per above teachings.

[0213] This is followed by the diagnosis phase in which a clinical biological sample 204 is obtained from an undiagnosed patient 202. Diagnostic platform 100, specifically its diagnosis module 124 now uses the reference models trained by learning module 120 to predict a labeled mitochondrial dysfunction or dysfunctions based on an unseen biomarker measurement generated from sample 204. The biomarker measurement is made by a measuring instrument, such as a high-resolution mass spectrometer (not shown in FIG. 6). The unseen biomarker measurement contains a mass spectrum of analytes and their quantities observed in sample 204 according to above teachings.

[0214] As noted above, diagnosis phase 56 of FIG. 1 carried out largely by diagnosis module 124 of FIG. 6 may only require a reduced set of analytes to be measured from sample 204 than targeting phase 54. This is because trained models 140A1-AN may be further refined during targeting to select a smaller number of features/analytes most predictive of a given mitochondrial dysfunction than during learning phase 52.

[0215] Diagnosis module 124 uses mapping 130 of mitochondrial dysfunctions and known mitochondrial diseases to diagnose the presence of one or more mitochondrial dysfunctions/diseases in patient 202 per above discussion. Thus, a key benefit of platform 100 is that it can diagnose the presence of a potential genetic mitochondrial disease(s) in patient 202 without the requirement of DNA sequencing. In other words, if patient 202 is predicted to have D3 which is mapped to Encephalomyopathy (see Table 8 above), then platform 100 can diagnose patient 202 of Encephalomyopathy without requiring DNA sequencing. This is an important innovation over the prevailing art. Platform is also useful for patients whose sequenced genetic data is not indicative of the mitochondrial disease characterized by the mitochondrial dysfunction(s).

[0216] Furthermore, platform 100 can also be used to issue therapeutic recommendations based on any known rescuers for the mitochondrial dysfunctions associated with the diagnosed disease in the patient. Recall from learning phase above that models 140A1-AN can also be used to produce a ranked list of predicted dysfunctions Dm in a given patient, such as patient 202 of FIG. 6. As a result, platform 100, specifically its diagnosis module 124 generates a multilevel diagnostic ranking 206 for patient 202 once the diagnosis phase is complete.

[0217] Specifically, diagnostic ranking 206 consists of columns including: the stressor/insult or mitochondrial inhibitor whose introduction had induced the predicted labeled mitochondrial dysfunctions(s), the top (for example 3) ranked predicted dysfunctions, associated/mapped genetic mitochondrial disease discovered during targeting, any known causal genetic defects for the genetic disease, and any rescuer compound that is known to ameliorate the conditions associated with the corresponding labeled dysfunction(s). As shown in Table 10, diagnostic ranking 206 is indicative of the correlations that exist between the originating mitochondrial inhibitor used in the experiments and the genetic defects or genetic patterns expressed by the undiagnosed patient. This knowledge alone is helpful in suggesting improved therapies for the patient.

[0218] A partial excerpt of an exemplary diagnostic ranking 206 for our patient 202 of FIG. 6 is provided below in Table 10. Note that in some embodiments, diagnostic ranking 206 may be built in parts, with some columns populated during learning, others during targeting and/or still others during diagnosis. Further, ranking 206 may also be stored in database 170 (now shown in FIG. 6) for later retrieval and analysis. In Table 10 below, R1 is the presumed rescuer compound for dysfunction D3 (Table 2 above) while there is presumed to be no available/known rescuer compound for dysfunction D10.

TABLE-US-00009 TABLE 10 Diagnostic Ranking 206 for patient 202 Weighted/ Cause/ Dys- Normalized Stressor/ Genetic Genetic Known Rank function Prediction Inhibitor Disease Defect Rescuer 1 D12 75% R8875 Complex I Autosomal Vitamin Deficiency E Hydro- quinone 2 D3 48% SML1122 Encephalo- Autosomal R1 myopathy Recessive 3 D10 21% O4876 Neuropathy, Mitochondrial N/A Ataxia, & DNA Retinitis point Pigmentosa mutations in (NARP) genes associated with Complex V: T8993G, (also T8993C by some research)

[0219] Thus, we observe from the diagnosis phase that the key benefits of the present design include: [0220] (1) Diagnosis of a potential genetic disease without requiring DNA sequencing. This diagnosis is obtained as the top-ranking dysfunction and associated genetic disease from diagnostic ranking 206 presented above. [0221] (2) Prognosis or potential vulnerability/risk of patient 202 to future development of genetic diseases based on the lower-ranked dysfunctions and associated genetic diseases from diagnostic ranking 206 presented above. Specifically, patient 202 may have a tendency to develop D3 and associated Encephalomyopathy, as well as D10 and associated NARP over time. [0222] (3) A ranked list of potential therapeutic recommendations based on any known rescuer compounds that are known to alleviate the mitochondrial dysfunction(s) predicted in the patient. At least, platform 100 may be used to narrow the range of potential diagnoses or to recommend a range of diagnoses for future study.

[0223] The above are important innovations of the present design over the prevailing art.

[0224] Note that there are laboratory processes that may or may not be necessarily computer-implemented. These may include drawing reference and target samples during the learning and targeting phases respectively, as well as drawing the clinical sample during the diagnosis phase from undiagnosed patient 202 above. These may also include operating mass spectrometer(s) to obtain corresponding biomarker measurements during these phases. That is why some embodiments within the present scope will presume such tasks to be under the purview of respective computing modules i.e. learning, targeting and diagnosis modules of platform 100 as shown in FIG. 6. However, other embodiments within the present scope may practice such tasks to be manual/mechanical and outside the purview of these computing modules.

[0225] As a result, learning phase 52 of protocol 50 of FIG. 1 which contains all aspects of learning until the development of the reference models 140A1-AN is said to be largely implemented by learning module 120 of FIG. 1, 6. Similarly, targeting phase 54 which contains all aspects until the development of mapping 130 is largely implemented by targeting module 122 and diagnosis phase 56 is largely implemented by diagnosis module 124 of FIG. 6.

[0226] Let us now consider another practical application of the present embodiments. Mitochondrial inhibitor Rotenone (D12 in Table 2 above) is known to function by knocking out Complex 1 in mitochondria, interfering with the cells ability to consume energy. By deploying learning phase 52 (see FIG. 1) on diagnostic platform 100 of FIG. 2, let us consider that learning models 140A1, 140A2 and 140A3 are trained to predict the presence of D12 based on reference biomarker measurements of in vitro samples 102 obtained from inhibited and uninhibited cultures per above teachings.

[0227] By deploying targeting phase 54 (see FIG. 1), diagnostic platform 100 is able to find association of dysfunction D12 with the known genetic mitochondrial disease(s) involving Complex 1 deficiency (Table 9 above). Now consider that during clinical or diagnosis phase 56 (see FIG. 1), based on a blood sample a patient 202 of FIG. 6 is predicted by models 140A1-A3 to have dysfunction D12 and the associated mitochondrial disease of Complex 1 deficiency per above teachings.

[0228] Since Rotenone is implicated in causing ferroptosis due to Complex 1 deficiency (see at least reference entitled Ferroptosis: An Iron-Dependent Form of Non-Apoptotic Cell Death, by Dixon et al. published in Cell 149(5): 1060-1072, doi:10.1016/j.cell.2012.03.042, dated May 25, 2012) and vitamin E hydroquinone has been discovered to be a potent inhibitor of ferroptosis, this knowledge of vitamin E hydroquinone as a rescuer of D12 may be maintained by platform 100 in a rescuer table of the form <Dysfunction Dm>, <Any known Rescuer> in database 170.

[0229] Other mitochondrial dysfunctions that are linked to ferroptosis of cells may also be maintained in the rescuer table along with vitamin E hydroquinone as the rescuer. In the same rescuer table, the presumed rescuer R1 of dysfunction D3 and a null/empty field indicating no known rescuer for dysfunction D10 may also be maintained. Rescuer table may be updated by a user of platform 100 or updated via a script or still other techniques known in the art.

[0230] As a result, platform 100 and specifically its diagnosis module 124 uses the above rescuer table to populate diagnostic ranking 206 with vitamin E hydroquinone as the rescuer compound for D12 for our patient 202. Similarly, it populates R1 as the rescuer for D3, and an empty field for the rescuer for D12 in diagnostic ranking 206 for patient 202. The Known Rescuer column of Table 10 is then used by platform 100 and/or an associated medical professional to incorporate in personalized targeted therapies for patient 202 that were heretofore unavailable. Such a capability is tremendous contribution of the present design over prevailing art.

[0231] Since mitochondrial dysfunctions are implicated in the causes of a large number of diseases, the present techniques may be employed in the diagnosis and treatment of such diseases. These diseases include at least neurodegenerative, cardiovascular, autoimmune, neurobehavioral, psychiatric, gastrointestinal and musculoskeletal diseases. These may also include types of diabetes, metabolic syndromes, fatiguing illnesses, cancers and chronic infections.

[0232] Consequently, a non-limiting list of neurodegenerative diseases for which mitochondrial dysfunctions may be predicted by the present techniques in order to improve potential treatments include Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis (ALS) and Friedreich's ataxia. Similarly, a non-limiting list of cardiovascular diseases for which mitochondrial dysfunctions may be predicted by the present techniques in order to improve potential treatments include a variety of vascular conditions including atherosclerosis. A non-limiting list of autoimmune diseases diagnosable and treatable by the present techniques include sclerosis, systemic lupus erythematosus and Type 1 diabetes.

[0233] In a similar manner, a non-limiting list of neurobehavioral diseases for which mitochondrial dysfunctions may be predicted by the present techniques in order to improve potential treatments include autism spectrum disorder, schizophrenia, a bipolar disorder, a mood disorder, depression, attention deficit hyperactivity disorder (ADHD) and post-traumatic stress disorder (PTSD). A non-limiting list of fatiguing illnesses for which mitochondrial dysfunctions may be predicted by the present techniques in order to improve potential treatments include chronic fatigue syndrome and Gulf War illness. A non-limiting list of musculoskeletal diseases diagnosable and treatable by the present techniques include fibromyalgia and skeletal muscle atrophy.

[0234] The above teachings are provided as reference to those skilled in the art in order to explain the salient aspects of the invention. It will be appreciated from the above disclosure that a range of variations on the above-described examples and embodiments may be practiced by the skilled artisan without departing from the scope of the invention(s) herein described. The scope of the invention should therefore be judged by the appended claims and their equivalents.