Fig. 1: Workflow of SVLearn. | Nature Communications

Fig. 1: Workflow of SVLearn.

From: SVLearn: a dual-reference machine learning approach enables accurate cross-species genotyping of structural variants

Fig. 1: Workflow of SVLearn.The alternative text for this image may have been generated using AI.

Based on a known SV set, an alternative (ALT) genome was constructed relative to the reference (REF) genome. Short reads were mapped to REF and ALT genomes to generate REF BAM and ALT BAM files, respectively. SV features were extracted from the two genomes. Alignment features and Paragraph features (optional) were extracted for each SV from the REF BAM and ALT BAM files. The true genotype (GT) of each SV is taken as the label used for training. The model then takes the feature matrix as input and outputs the predicted genotypes of SVs used.

Back to article page