Abstract: Disclosed herein is a method for detecting deletion in a gene sequence. The method comprises receiving, by a processor, training sequencing data, which comprises multiple training reads associated with gene sequences with deletion and gene sequences without deletion. The processor splits each of the multiple training reads into multiple training segments shorter than the training reads and trains a machine learning model on the multiple segments. The processor receives testing sequencing data comprising multiple testing reads, splits each of the multiple testing reads into multiple testing segments, and evaluates the trained machine learning model to the multiple testing segments to detect deletion in the testing sequencing data. No alignment or variant calling is necessary, which reduces the computational complexity of the evaluation step significantly.

Abstract: This disclosure relates to computationally efficient processing of sequencing data relating to amyotrophic lateral sclerosis (ALS). A processor receives unaligned training reads and determines training sub-sequences from them. The processor then counts the training sub-sequences in a control group and in a group diagnosed with ALS and determines a measure of change, for each of the training sub-sequences, in the counting between the control group and the group with ALS. The processor further selects a subset of training sub-sequences that are distal from a mean value of the measure of change and then receives testing sequencing data comprising multiple unaligned testing reads. The processor determines sub-sequences from the testing reads, counts the sub-sequences that are in the subset, and determines a diagnostic output value related to ALS for the sample based on the counting of the testing sub-sequences that are in the subset.