Epsilon Archive for Student Projects

Abstract

One of the most important challenges in human medicine is the identification of genetic variants underlying complex diseases but we are still not able to assess the risk of disease by genetic profiling. Also, as the complete functional properties remain unknown for most genes it is very important to further functionally annotate genomes. The large abundance of domestic animals kept by humans constitute a valuable asset for the identification of genetic elements involved in complex disease and to unravel gene functions. Advantages of using animal models in studies of complex traits include the possibilities to strictly control environmental variation and to perform invasive measurements. Domestic animals have the additional advantage that different breeds segregate for traits of importance to human diseases.

Massive parallel sequencing technology (NextGen) enables researchers to resequence already completed genomes and enables whole-genome sequence information from multiple DNA samples to be determined in a short time. This makes it possible to, in a global manner, determine the extent of sequence polymorphisms among members within a species. Insertions and deletions are among the structural variation events that may be responsible for the variation in the expression of traits related to growth and reproduction in chickens. Rubin et al. (2010) pioneered the use of NextGen sequencing for population-based studies. The current study presented here will now extend previous analysis performed by Rubin et al., 2010 using paired reads, which makes identification of insertions/deletions in the chicken genome possible. Chicken lines whose sequences have been analyzed in this study include the red junglefowl as a reference bird, the High growth line (HL), the low growth line (LL) both established from White Plymouth Rock chickens in 1957 (Dunnington and Siegel, 1996) and the White Leghorn line L13 (WL_L13). Whole genome resequencing was performed on three lines of chickens where genomic DNA from 11 chicken individuals were pooled from each line to generate paired-end reads. Paired-end reads of 50 bp each were generated using AB-SOLiDTM v3.0. The overall coverage of the reads was 20-25x. Most deletions were found for sizes ranging from 4kbp to 10kbp in the HL, LL and the WL_L13 and most insertions were detected for size ranging from 1.1kbp to 3.1kbp. The coverage from matching pipeline was performed on chromosome 13 that contains 326 regions detected as deletions using mate-pair reads. The identified deletions overlapping duplicated regions in the chicken genome were removed for chromosome 13 in HL and considering a median coverage equal to zero (of reads mapped by the matching pipeline) in putative deleted regions detected using paired reads resulted in 73 deleted regions. It is possible that most of the identified deletions are false positives and further refinement in the approach is required to increase the probability of finding only true deletions and insertions as well. This could be due to the artifacts in the library as well as the in the assembly of the sequenced chicken genome. Further analysis can be carried out to find the exact breakpoints of the deletions. We suggest the approach of taking the set of reads that are unmapped in the genome and align them again by splitting those into two parts and allowing a gap or an insertion between them, as reads spanning these features should be present in the genomes of populations bearing true deletions and insertions, respectively.