Using deep learning to identify recent positive selection in malaria parasite sequence data

• Published in: Malaria journal 2021-06, Vol.20 (1), p.270-270, Article 270
• Main Authors: Deelder, Wouter, Benavente, Ernest Diez, Phelan, Jody, Manko, Emilia, Campino, Susana, Palla, Luigi, Clark, Taane G
• Format: Article
• Language: English
• Subjects: Accuracy; Algorithms; Data; Datasets; Decision making; Deep learning; Deep Learning - statistics & numerical data; Development and progression; DNA; DNA sequencing; Drug resistance; Drug therapy; Drugs; Gene mutations; Genes; Genetic aspects; Genome Size; Genome, Protozoan; Genomes; Haplotypes; Health aspects; Homozygosity; Human diseases; Identification and classification; Learning algorithms; Machine learning; Malaria; Methods; Mutation; Neural networks; Nucleotide sequence; Parasites; Pathogens; Plasmodium; Plasmodium falciparum; Plasmodium falciparum - genetics; Plasmodium vivax; Plasmodium vivax - genetics; Population; Population genomics; Positive selection; Public health; Quantitative trait loci; Regions; Selection, Genetic; Sequence Analysis, DNA; Software; Vector-borne diseases; Whole genome sequencing
• ISSN: 1475-2875; 1475-2875
• DOI: 10.1186/s12936-021-03788-x

Malaria, caused by Plasmodium parasites, is a major global public health problem. To assist an understanding of malaria pathogenesis, including drug resistance, there is a need for the timely detection of underlying genetic mutations and their spread. With the increasing use of whole-genome sequencing (WGS) of Plasmodium DNA, the potential of deep learning models to detect loci under recent positive selection, historically signals of drug resistance, was evaluated. A deep learning-based approach (called "DeepSweep") was developed, which can be trained on haplotypic images from genetic regions with known sweeps, to identify loci under positive selection. DeepSweep software is available from https://github.com/WDee/Deepsweep . Using simulated genomic data, DeepSweep could detect recent sweeps with high predictive accuracy (areas under ROC curve > 0.95). DeepSweep was applied to Plasmodium falciparum (n = 1125; genome size 23 Mbp) and Plasmodium vivax (n = 368; genome size 29 Mbp) WGS data, and the genes identified overlapped with two established extended haplotype homozygosity methods (within-population iHS, across-population Rsb) (~ 60-75% overlap of hits at P