Disclosed herein are methods for improving detection and monitoring of human diseases. The methods can be used to provide spatial and/or developmental localization of the source of each differential mutation within the body. The methods can also be used to generate a mutation map of a subject. And the mutation map can be used to monitoring state(s) of health of one or more tissues of a subject.

This disclosure describes techniques for generating a visual path through a metabolic network, techniques include receiving reaction data for a starting metabolite, the reaction data comprising one or more potential reaction steps of the starting metabolite with one or more neighbor metabolites, generating from the one or more neighbor metabolites, a precursor metabolite list of precursor metabolites and a successor metabolite list of successor metabolites based on the reaction data, generating a user interface comprising the visual path that includes interactive visual data, the interactive visual data indicating each metabolite from the precursor metabolite list and the successor metabolite list, wherein each metabolite from the precursor metabolite list and the successor metabolite list is selectable and indicates a corresponding potential reaction step of the one or more potential reaction steps of the starting metabolite, and outputting the user interface for display at a display device.

A blood cell analysis method and a blood cell analyzer are provided. In the method and analyzer, characteristic information of white blood cell fragments is obtained based on side scattered light information and fluorescence information, characteristic information of platelets is obtained based on forward scattered light information and fluorescence information and then a count value for the platelets is acquired based on the characteristic information of the platelets and the characteristic information of the white blood cell fragments. The present invention can avoid the influence of the white blood cell fragments on the platelet counting, thereby ensuring the accuracy of the platelet counting without increasing costs.

A blood cell analysis method and a blood cell analyzer are provided. In the method and analyzer, characteristic information of white blood cell fragments is obtained based on side scattered light information and fluorescence information, characteristic information of platelets is obtained based on forward scattered light information and fluorescence information and then a count value for the platelets is acquired based on the characteristic information of the platelets and the characteristic information of the white blood cell fragments. The present invention can avoid the influence of the white blood cell fragments on the platelet counting, thereby ensuring the accuracy of the platelet counting without increasing costs.

A method for storing, by a processor, a genomic graph representing a plurality of individual genomes, including: storing a linear representation of a reference genome in a data storage; receiving a first genome; identifying variations in the first genome from the reference genome; generating graph edges for each variation in the first genome from the reference genome; generating for each generated graph edge: an edge identifier that uniquely identifies the current edge in the genome graph; a start edge identifier that identifies the edge from which the current edge branches out; a start position that indicates the position on the start edge that serves as an anchoring point for the current edge; an end edge identifier that identifies the edge into which the current edge joins in; an end position that indicates the position on the end edge that serves as an anchoring point for the current edge; and a sequence indicating the nucleotide sequence of the current edge; and storing the edge identifier, start edge identifier, start position, end edge identifier, end edge position, and sequence for each generated graph edge in the data storage. Based on this genome graph data structure, we further propose a scheme for specifying a path, which may traverse one or more edges, and the ways to extend existing genomic data formats such as SAM, VCF and MPEG-G to support the use of genome graph reference using our proposed coordinate system.

A method for storing, by a processor, a genomic graph representing a plurality of individual genomes, including: storing a linear representation of a reference genome in a data storage; receiving a first genome; identifying variations in the first genome from the reference genome; generating graph edges for each variation in the first genome from the reference genome; generating for each generated graph edge: an edge identifier that uniquely identifies the current edge in the genome graph; a start edge identifier that identifies the edge from which the current edge branches out; a start position that indicates the position on the start edge that serves as an anchoring point for the current edge; an end edge identifier that identifies the edge into which the current edge joins in; an end position that indicates the position on the end edge that serves as an anchoring point for the current edge; and a sequence indicating the nucleotide sequence of the current edge; and storing the edge identifier, start edge identifier, start position, end edge identifier, end edge position, and sequence for each generated graph edge in the data storage. Based on this genome graph data structure, we further propose a scheme for specifying a path, which may traverse one or more edges, and the ways to extend existing genomic data formats such as SAM, VCF and MPEG-G to support the use of genome graph reference using our proposed coordinate system.

A system and method for optimizing the latent space in generative machine learning models, and applications of the optimizations for use in the de novo generation of molecules for both ligand-based and pocket-based generation. The ligand-based optimizations comprise a tunable reward system based on a multi-property model and further define new measurable metrics: molecular novelty and uniqueness. The pocket-based optimizations comprise an initial multi-property optimization followed up by either a seed-based optimization or a relaxed-based optimization.

A system and method for optimizing the latent space in generative machine learning models, and applications of the optimizations for use in the de novo generation of molecules for both ligand-based and pocket-based generation. The ligand-based optimizations comprise a tunable reward system based on a multi-property model and further define new measurable metrics: molecular novelty and uniqueness. The pocket-based optimizations comprise an initial multi-property optimization followed up by either a seed-based optimization or a relaxed-based optimization.

There is a need for more effective and efficient predictive data analysis solutions and/or more effective and efficient solutions for generating image representations of genetic variant data. In one example, embodiments comprise receiving an input feature, generating one or more image representations of the input feature, generating a tensor representation of the one or more image representations, generating a plurality of positional encoding maps, generating an image-based prediction based at least in part on the image representation, and performing one or more prediction-based actions based at least in part on the image-based prediction.

There is a need for more effective and efficient predictive data analysis solutions and/or more effective and efficient solutions for generating image representations of genetic variant data. In one example, embodiments comprise receiving an input feature, generating one or more image representations of the input feature, generating a tensor representation of the one or more image representations, generating a plurality of positional encoding maps, generating an image-based prediction based at least in part on the image representation, and performing one or more prediction-based actions based at least in part on the image-based prediction.