Case Study of Sequence Capture Enrichment Technology: Identification of Variation Underpinning Developmental Syndromes in an Amniote Model

Abstract

Chicken developmental mutants are valuable for discovering sequences and pathways controlling amniote development. Herein we applied the advanced technologies of targeted sequence genomic capture enrichment and next-generation sequencing to discover the causative element for three inherited mutations affecting craniofacial, limb and/or organ development. Since the mutations (coloboma, diplopodia-1 and wingless-2) were bred into a congenic line series and previously mapped to different chromosomes, each targeted mutant causative region could be compared to that of the other two congenic partners, thereby providing internal controls on a single array. Of the ~73 million 50-bp sequence reads, ~76% were specific to the enriched targeted regions with an average target coverage of 132-fold. Analysis of the three targeted regions (2.06 Mb combined) identified line-specific single nucleotide polymorphism (SNPs) and micro (1–3 nt) indels. Sequence content for regions indicated as gaps in the reference genome was generated, thus contributing to its refinement. Additionally, Mauve alignments were constructed and indicated putative chromosomal rearrangements. This is the first report of targeted capture array technology in an avian species, the chicken, an important vertebrate model; the work highlights the utility of employing advanced technologies in an organism with only a “draft stage” reference genome sequence.

Defining the Sequence Elements and Candidate Genes for the Coloboma Mutation

“…A subset of those samples listed above [n = 28 mutant (Z co /W) and 2 control (Z + /Z + ) individuals] was used in the variant validation portion of this study (see below). To test for the putative translocation event (described in [32]), 2 control (Z + /Z + ), 4 heterozygous (Z + /Z co ), and 4 mutant (Z co /W; which included the two samples sequenced with the CA/NGS) samples from the archived coloboma DNA were employed.…”

“…Two samples, co-275F and co-276F (superscripts H and G , respectively), were utilized in the CA/NGS technology [32]. Note that co-275F has a minimum and maximum causative region (CR min and CR max ) of 494,598 and 994,599 bp, respectively.…”

“…990 kb) coloboma candidate region on GGA Z (20,813,939-21,804,206 (November 2011, galGal4 assembly) formerly reported as 994,523 bp; (GGA Z: 20,368,747-21,363,270 (May 2006, galGal3 assembly))) [32] as part of a three-mutant sequencing screen. As described by Robb and Delany [32], mutant-specific DNA pools (50 µg total) were sent to SeqWright DNA Technology Services (Houston, TX) for barcoding followed by library production (base pair peak of 150–200), targeted enrichment (for chromosomes 1, 12, and the Z) and SOLiD ™ V3 Plus Platform sequencing (50bp, single-end sequencing) (Applied Biosystems, Foster City, CA). A total of 3.64 Gbp of data were generated using this technology.…”

“…To refine and reduce the number of potential coloboma - causative elements, multiple pairwise-line comparative genomic analyses were utilized (described in [32]). Shared variants were submitted to NCBI (Accessions: ss472340674–ss472343089) and were therein eliminated as possible candidates whereas any variant unique to coloboma alone (i.e., not found within control lines or not previously reported in NCBI and the UCSC genome browsers) was further studied here.…”

“…That is, a targeted genomic capture enrichment technology (a.k.a. capture array (CA)) was employed to identify the element causing this mutation [32]. Here we focus on the CA/NGS analysis of the genetic features discovered (e.g., SNPs, insertions, deletions) which include normal polymorphisms of the introgressed region as well as the mutant-specific variant element.…”