Systems and methods for correcting for noise and systemic variations in sequencing data

Abstract

The disclosure generally pertains to providing an accurate genotype for high repeat genes. First, repeat structures of sequencing reads are identified and sequence reads with abnormal repeat structures are removed to generate a first dataset. The sequence reads of the first dataset are filtered to exclude sequences with structural artifacts to generate a second dataset. Data from the second dataset is input to a model to generate a third dataset, the third dataset including sequencing reads categorized by allele. Next, a consensus sequence is generated for each allele in the third dataset to generate a fourth dataset of consensus reads for all alleles in the third dataset. Each allele of the fourth dataset is translated to generate a fifth dataset, the fifth dataset including the corresponding amino-acid sequence of each allele. A protein-altering variant status is then determined based on the amino-acid sequences of the fifth dataset.