GECKO is a genetic algorithm to classify and explore high throughput sequencing data

Abstract

Comparative analysis of high throughput sequencing data between multiple conditions often involves mapping of sequencing reads to a reference and downstream bioinformatics analyses. Both of these steps may introduce heavy bias and potential data loss. This is especially true in studies where patient transcriptomes or genomes may vary from their references, such as in cancer. Here we describe a novel approach and associated software that makes use of advances in genetic algorithms and feature selection to comprehensively explore massive volumes of sequencing data to classify and discover new sequences of interest without a mapping step and without intensive use of specialized bioinformatics pipelines. We demonstrate that our approach called GECKO for GEnetic Classification using k-mer Optimization is effective at classifying and extracting meaningful sequences from multiple types of sequencing approaches including mRNA, microRNA, and DNA methylome data.

iMOKA: k-mer based software to analyze large collections of sequencing data

“…iMOKA imports sequencing files in FASTQ, FASTA, BAM format, or SRR identifiers via its user interface. It then counts the occurrences of all sequences of given length k (default 31) [ 9 ] using the KMC3 software [ 10 ] in each sample (Fig. 1 ).…”

“…The algorithms we benchmarked were DESeq2 [ 12 ], edgeR [ 13 ], and limmaVoom [ 14 ] for TPM and sequencing counts; iMOKA for k -mer counts; and Whippet [ 15 ] for alternative splice site usage. We excluded four other k -mer based methods HAWK [ 16 ], KOVER [ 17 ], Kissplice [ 11 ], and GECKO [ 9 ] because they were respectively impossible to run on such big datasets due to segmentation fault errors, were unable to find k -mers that could classify the input samples or, for the last two methods, were killed after 2 weeks of runtime on our computer cluster.…”

“…For each sample, KMC3 [9] is used to count the k-mers of the length chosen by the user (default k = 31). Its output is converted into a sorted binary file optimized for the following steps of iMOKA and a JSON file containing the metadata information.…”

“…However, when translated to mammalian genomes, k-mer representation results in a k-mer count matrix with as many columns as there are samples and as many rows as there are k-mers, generally billions. Exploring such large matrices to find biologically relevant k-mers is intractable unless the analysis focuses only on a very small subset of the sequencing data [5] or by using metaheuristics that provide partial solutions [9].…”

A Hybrid Heuristic Dimensionality Reduction Methods for Classifying Malaria Vector Gene Expression Data

“…Generally, feature extraction procedures can be used to efficiently aid feature selection [35], [39], [49], [50], by using feature selection to select the original gene subset or benefit from eliminating redundant genes. Combination of numerous feature extraction techniques can be applied to extract the initial feature subsets [57]- [60].…”