SNPfiltR: an R package for interactive and reproducible SNP filtering

Abstract

Here I describe the novel R package SNPfiltR and demonstrate its functionalities as the backbone of a customizable, reproducible SNP filtering pipeline implemented exclusively via the widely adopted R programming language. SNPfiltR extends existing SNP filtering functionalities by automating the visualization of key parameters such as depth, quality, and missing data, then allowing users to set filters based on optimized thresholds, all within a single, cohesive working environment. All SNPfiltR functions require a vcfR object as input, which can be easily generated by reading a SNP dataset stored as a standard vcf file into an R working environment using the function read.vcfR() from the R package vcfR. Performance benchmarking reveals that for moderately sized SNP datasets (up to 50M genotypes with associated quality information), SNPfiltR performs filtering with comparable efficiency to current state of the art command-line-based programs. These benchmarking results indicate that for most reduced-representation genomic datasets, SNPfiltR is an ideal choice for investigating, visualizing, and filtering SNPs as part of a cohesive and easily documentable bioinformatic pipeline. The SNPfiltR package can be downloaded from CRAN with the command [install.packages(“SNPfiltR”)], and a development version is available from GitHub at: (github.com/DevonDeRaad/SNPfiltR). Additionally, thorough documentation for SNPfiltR, including multiple comprehensive vignettes, is available at the website: (devonderaad.github.io/SNPfiltR/).

Easy-to-use R functions to separate reduced-representation genomic datasets into sex-linked and autosomal loci, and conduct sex-assignment

“…This has the advantage of taking into account random sampling and genotyping errors that affect loci differently. In fact, this approach is available in VCFtools but not yet in dartR , snpR orSNPfiltR (Hohenlohe et al 2011;Denecek et al 2011;Mijangos et al 2022;Hemstrom & Jones 2022;DeRaad 2022). Nonetheless, we would like to emphasize that this is not a Hardy-Weinberg equilibrium filter (which requires critical thinking to be correctly applied and interpreted; Waples 2015), and should be used only when looking to obtain neutral autosomal loci (cf.…”

Drivers of phenotypic divergence in a Mesoamerican highland bird

“…For the final alignment, we retained loci that were present in 75% of the samples. For analyses that require unlinked loci, we further filtered our dataset using the R packages SNPfiltR ( DeRaad, 2021 ) and vcfR ( Knaus & Grünwald, 2017 ), retaining only loci that were more than 1000bp away from one another.…”

The maintenance of self-incompatibility: overdominance of inbreeding depression and S-linked genetic load

“…GBS products from separate genomic regions that map to such short, artifactual contigs would be expected to yield heterozygote genotype calls. To minimize these artifacts we: (1) filtered out heterozygotes when allele balance fell below 0.333 or above 0.667 using the R- package SNPfiltr v. 1.00 (DeRaad, 2022), (2) restricted our analysis to SNPs that map to scaffolds > 100 kbp in length (these account for 73% of the L. alabamica assembly); (3) filtered out SNP loci that exhibited only heterozygotes when adjacent SNP loci on the same scaffold exhibited segregation ratios that were not significantly different from 1:2:1 by Chi-square tests, on the basis that evidence of strong selection should be shared by adjacent SNP loci; and (4) restricted our analysis to loci where data were available from > 90% of the progeny of a cross since genotyping errors are more likely to have occurred when few progeny per family are genotyped successfully.…”