Deep sequencing strategies for mapping and identifying mutations from genetic screens

Abstract

The development of next-generation sequencing technologies has enabled rapid and cost effective whole genome sequencing. This technology has allowed researchers to shortcut time-consuming and laborious methods used to identify nucleotide mutations in forward genetic screens in model organisms. However, causal mutations must still be mapped to a region of the genome so as to aid in their identification. This can be achieved simultaneously with deep sequencing through various methods. Here we discuss alternative deep sequencing strategies for simultaneously mapping and identifying causal mutations in Caenorhabditis elegans from mutagenesis screens. Focusing on practical considerations, such as the particular mutant phenotype obtained, this review aims to aid the reader in choosing which strategy to adopt to successfully clone their mutant.

Keywords: C. elegans; EMS; NGS; SNP mapping; WGS; deep sequencing; forward genetics; genetic mapping; next-generation sequencing; whole genome sequencing.

Figure 1. Principle of each of the simultaneous mapping and mutant identification methods used in C. elegans. (A) SNP-based strategy requiring crossing of the mutant (N2 background) to CB4856, pooling of multiple isolated F₂ recombinants that display the phenotype and deep sequencing to discover a genomic region where only N2 SNPs reside. This region should contain the causal mutation (green star). Adapted from reference , (B) EMS-based strategy whereby the mutant is backcrossed multiple times until EMS-induced mutations (white stars) are cleared, except for those genetically linked to the causal mutation (green star). The cluster of EMS-induced GA > CT nucleotide changes is detected by deep sequencing and localized to a genomic region that should contain the causal mutation. (C) A bulk-segregant approach to the EMS-based mapping strategy that involves a single backcross followed by pooling of F₂ recombinants and deep sequencing to identify a genomic region where EMS induced GA > CT are exclusive (white stars). In this region, the causal mutation (green star) should be located.

Figure 2. Flowchart of SNP-based simultaneous mapping and mutant identification using deep sequencing. This strategy is similar to traditional SNP mapping techniques performed before whole genome sequencing was feasible. However, instead of analyzing the linkage of SNPs to a causal mutation one at a time through laborious identification procedures, deep sequencing is used to interrogate most of them simultaneously (see ref. 21). Twenty to 50 F₂ progeny displaying the mutant phenotype are picked from a cross between the mutant (usually in a Bristol N2 background) and a polymorphic strain such as CB4856. They are then singled onto fresh plates, grown and pooled before DNA is extracted and sequenced. From the deep sequencing data and with the use of available software, a genomic region linked to the causal mutation will be identified and nucleotide mutations within this region will also be revealed. A list of candidate causal mutations that affect genes within this region will also be assembled. The original mutant is then subjected to multiple rounds of backcrossing or outcrossing (≥ six simple backcrosses or equivalent recommended) to remove all other EMS-induced mutations from its background and then tested via several methods (e.g., complementation analysis and transgenic rescue) to determine which candidate mutation is causing the phenotype.

Figure 3. Flowchart of EMS-based simultaneous mapping and mutant identification using deep sequencing. This strategy for identifying a causal mutation traces the distinctive changes in nucleotides caused by EMS itself to map the location of the mutation. Several rounds of backcrossing remove EMS-induced nucleotide changes that are un-linked to the causal mutation, while leaving a “hot spot” of EMS damage surrounding the causal mutation (see ref. 12). Only a single recombinant is needed at the end of backcrossing to grow and extract DNA for deep sequencing. This is also advantageous when working with a mutant with a low penetrance or subtle phenotype. Deep sequencing data and analysis with available software will reveal a genomic region linked to the causal mutation. Nucleotide mutations within this region will also be revealed and a list of candidate causal mutations that affect genes within this region will be assembled. The mutant can then be analyzed directly (since it has already been backcrossed) via several methods (e.g., complementation analysis and transgenic rescue) to determine which candidate mutation is causing the phenotype.

Figure 4. Flowchart of a bulk segregation approach to EMS-based mapping using deep sequencing. Only a single backcross would be necessary to perform EMS-based mapping if multiple recombinant F₂ mutant progeny were singled, grown and pooled for deep sequencing. This approach has the added advantage of saving time later by reducing the number of backcrosses needed before mutant analysis as well as avoiding a CB4856 cross, which may preclude some mutant phenotypes. More than one backcross can be performed prior to selection of 20–50 recombinants, which would further increase mapping accuracy. Again, deep sequencing data and analysis with available software will reveal a genomic region linked to the causal mutation. Nucleotide mutations within this region will also be revealed and a list of candidate causal mutations that affect genes within this region will be assembled. The mutant is then be backcrossed, although a less number of times since it has already undergone a round of backcrossing, and then tested via several methods (e.g., complementation analysis and transgenic rescue) to determine which candidate mutation is causing the phenotype.