Sequencing technologies have made it possible to carry out whole genome sequencing (WGS) to identify the genetic variants present in mutant organisms. In addition, methods have been developed that simultaneously facilitate mapping of the causal mutation. Currently, two principal strategies for "mapping by sequencing" are used by C elegans researchers. In one approach, mutants are crossed to the polymorphic HA strain, which provides extensive SNP mapping information and localization of the causal mutation. This approach, however, is not highly practical for complex genetic backgrounds, such as the case with genetic suppressors, and the numerous polymorphisms present in the HA strain can alter phenotypes in ways that are unpredictable. A second method, EMS density mapping, exploits of the higher concentration of signature-EMS alterations in regions surrounding causal mutations following multiple backcrosses. The resolution of EMS density mapping, however, can be quite low and this method is not applicable for alleles that arise spontaneously or through other methods. We recently obtained ~30 strong suppressors
nekl-2;
nekl-3 double-mutant larval lethality. Our screens included a novel strategy using counter-selection that was enabled by the
peel-1 toxin. Notably, neither the
nekl-2 or
nekl-3 single mutants used for this screen display phenotypes on their own, nor did the majority of obtained suppressors. In order to identify the causal mutations that suppress
nekl-2;
nekl-3 lethality, we designed and implemented a novel refinement strategy we call the Sibling Subtraction Method (SSM). For our approach, suppressed strains were backcrossed several times to ensure Mendelian inheritance and to partially remove background mutations. While performing the final backcross, we isolated strains that were homozygous for the suppressor mutation as well as strains that were non-suppressed (i.e, homozygous wild type at the suppressor loci). These isolates were combined to create a Suppressed DNA pool and a Non-suppressed Comparitor DNA pool, respectively. Following WGS, alignment, and filtering steps, common variants present in both the Suppressed and Comparitor pools were subtracted. This resulted in a very small number of candidate variants (~5) that caused non-synonymous coding changes in each of the suppressor strains. Notably, we were able to subtract more than 95% of background variants using this method. The bioinformatics analysis was performed using Cloudmap workflows in the Galaxy web platform, which are freely available and widely used by the C elegans field. Our strategy was validated on five suppressor strains and, most importantly, is generally applicable for all varieties of mutations and genetic backgrounds. As such, SSM largely obviates the need to use the HA strain and provides better candidate refinement than EMS density mapping methods.