Abstract: This disclosure describes methods, non-transitory computer readable media, and systems that can utilize a machine-learning model to refine structural variant calls of a call generation model. For example, the disclosed systems can train and utilize a structural variant refinement machine-learning model to reduce false positives and/or false negatives. Indeed, the disclosed systems can improve or refine structural variant calls (e.g., between 50-200 base pairs in length) determined by a call generation model by training and utilizing the structural variant refinement machine-learning model. As disclosed, the systems can determine sequencing metrics and can customize training data for a structural variant refinement machine-learning model to generate modified structural variant calls.

Abstract: It is an invention aimed at providing methods for optimizing read information mapped to the particular gene loci such as MHC loci under a framework of probabilistic statistical processing. In the present invention, a step in which for all reads, calculation of the expected number of mappings to alleles of the particular gene loci is performed for read information in which mapping of reads to alleles of the particular gene loci is identified, a step in which the total number of expected mappings for each allele is calculated, and a step in which the fraction of reads allocated to each allele is calculated, are repeatedly executed, in which the optimization of read information is performed in a computer, and based on the optimized information, a method of easily and accurately estimating the genotypes of the particular gene loci, a computer system, and a computer program capable of easily and accurately estimating the genotype of the particular gene loci, are provided.

Abstract: The present invention provides a selection method of tag SNPs, for constituting a group of nucleic acid probes corresponding to the tag SNPs, the tag SNPs being used for performing imputation of information on SNPs of human genome by using human genome information, the human genome information including information on a group of SNPs, the genotypes of the SNPs being identified in multiple individuals, in which method a sum of mutual informations between tag SNP candidates and target SNPs is used as an index for selecting the tag SNPs, and a computer system based on the principle, a computer program, and a DNA microarray on which nucleic probes corresponding to the tag SNPs selected by the means, and a production method thereof.