Evolution, Molecular

We determined and vetted the whole genome sequences of 84 + 1 inbred Drosophila lines representing five global populations (Greenberg et al. 2010 (link); the “+1” is for line ZW184, recovered in Zimbabwe but appearing to be a very recent migrant, and so is left out of population structure studies). An expanded version of our Materials and Methods section is provided as Supporting Information, File S1. To summarize, genomic DNA was extracted from pools of 50 adult females, and we generated paired-end 100 nt Illumina reads at average 12.5× depth for each line (File S1, part 2). Sequence reads were aligned with the Burrows-Wheeler Alignment Tool (BWA, Li and Durbin 2009 (link); File S1, part 3) to the reference D. melanogaster genome, and SNPs and small indel genetic variants were called using the Genome Analysis Tool Kit (GATK; McKenna et al. 2010 (link); Depristo et al. 2011 (link); File S1, part 4). Genetic variant calls were validated in two ways, by resequencing one line to 100× depth and by double-digest restriction-site associated (ddRAD; Peterson et al. 2012 (link)) resequencing of a consistent subset of the genome for 12 lines, which informed additional filtering to create the final variant call sets (File S1, part 6). We noted that some subregions of the genome in most lines had an unexpectedly high frequency of heterozygous SNP genotypes that validated at high frequency, which we define as “heterozygous blocks” (File S1, part 5). Finally, we investigated the whole-genome sequence dataset for large chromosomal inversions using a custom bioinformatics pipeline (Cardoso-Moreira et al. 2012 (link); M. Cardoso-Moreira, J. R. Arguello, D. Riccardi, S. Gottipati, J. K. Grenier, and A. G. Clark, unpublished data) that uses several available tools for genome mapping and structural variation detection [Novoalign (www.novocraft.com), Mosaik (Lee et al. 2014 (link)), Delly (Rausch et al. 2012 (link)), and BLAT (Kent 2002 (link)); File S1, parts 3 and 10]. Candidate inversions were validated by polymerase chain reaction (PCR) across at least one of the inversion breakpoints, and the genotype of each line was determined by looking for reads supporting the presence of the inversion breakpoint and/or the reference sequence bridging the breakpoint (File S1, part 11). Alignment files and final genetic variant genotypes (vcf files), as well as companion files including 1) het blocks per line, 2) genome callability (File S1, part 9), and 3) regions of genetic identity by descent (IBD; File S1, part 8), are described in Table S1.
Preliminary molecular evolution and population genetic analyses were carried out to provide additional data quality checks, as well as to provide initial broad characterizations of inter- and intrapopulation variation (File S1, parts 13-16). To summarize, our divergence analyses were based on an updated five-species whole-genome alignment that we generated [D. melanogaster (dm3), D. simulans (droSim2), D. sechellia (droSec1), D. erecta (droEre2), and D. yakuba (droYak2)], within which we include the recently improved D. simulans assembly (Hu et al. 2013 (link)). Population genetic analyses were carried out with the final SNP calls, but were further masked based on variant callability and regions of genetic IBD. Several analyses use an additional SNP subset, referred to as “neutral” SNPs, that fall within small introns or in fourfold degenerate coding positions as determined by our SNPeff annotation (Cingolani et al. 2012 (link)).

The DNA sequences were analysed using the sequence analysis software 5 and then blasted on to the NCBI sequence database to confirm the k13 propeller gene sequence identity by using the Basic Local Alignment Search Tool (BLAST) at http://blast.ncbi.nlm.nih.gov/Blast.cgi. The sequences were exported to bio edit sequence alignment editor 7.2.5 for manual editing and then in to MEGA 5 software version 5.10 for detection of polymorphism using the PF3D7_1343700 and K13-propeller gene sequences present in the NCBI database were used as the reference sequence. Additional single-nucleotide polymorphism (SNPs) analysis within the K13 propeller gene was performed using the DnaSP software version 5.10.01. To assess the selection pressure in P. falciparum parasite population in Kisii County, Tajima’ D statistic and Fu & Li’s D test in DnaSP software 5.10.01 were used. In this analysis, the study evaluated whether the P. falciparum k13 propeller domain sequence data show evidence of deviation from the neutrality theory of molecular evolution. The analysis was done using commands in the DnaSP software. In the DnaSP software, the probability of Tajima’s D and Fu & Li’s D are estimated by simulation. The test uses information on the frequency of mutations (allelic variation) [23 (link)]. Tajima’s D and Fu & Li’s D test is based on the fact that under the neutral model, estimates of the number of polymorphic sites and the average number of nucleotide differences are correlated. The critical values (Tajima’s D and Fu & Li’s D) obtained were used in interpreting the findings under the neutrality assumption [24 (link)].

Reads obtained from the sequencing machines included raw reads containing adapters or low-quality bases, which would affect the following assembly and analysis. The clean reads were retrieved after trimming adapter sequences and removal of low quality (containing >50% bases with a Phred quality score < 20) using the FastQC tool. Transcriptome de novo assembly was performed with the short reads assembling program-Trinity (Grabherr et al., 2011 (link)). Firstly, a short sequence library of K-mer length was constructed using high-quality sequences. Then the short sequence was extended by the overlap of K-mer-1 length between short sequences to obtain the preliminary spliced contig sequences. Next, Chrysalis clusters related contigs that correspond to portions of alternatively spliced transcripts or otherwise unique portions of paralogous genes and then builds Bruijn graphs for each cluster of related contigs. Finally, these Bruijn graphs were processed to find the path based on the reads and paired reads in the graphs to obtain the transcripts. To comprehensively obtain gene annotation information, genes were compared with six databases, including NR (NCBI non-redundant protein sequences), Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genome (KEGG), eggNOG (evolutionary genealogy of genes: Non-supervised Orthologous Groups), Swiss-Prot, and Pfam, and the annotation situation of each database was counted.