Quantifying Allele-Specific Gene Expression Using Personalized Genomes

• Activity Number: 440
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): Narayanan Raghupathy*+ and Kwangbom Choi and Steven C. Munger and Gary Churchill
• Companies: Jackson Laboratory and Jackson Laboratory and Jackson Laboratory and Jackson Laboratory
• Keywords: RNA-seq ; Alele-specific gene expression ; Gene Expression ; Personal Genome ; Alignment Bias

Abstract

Aligning short sequencing reads to a common reference genome is the first step in RNA-seq analysis. Genetic variations present in the sample, but not in the reference genome can lead to misalignments and incorrect expression estimates. Large-scale sequencing efforts have characterized millions of common genetic variants across human populations. However, developments of tools that can effectively utilize the individual-specific variation to inform expression quantitation lag behind. Current approaches employ two steps to quantify gene expression and allele-specific expression (ASE); gene expression is estimated from all alignments, while ASE is assessed separately by using only reads that overlap known SNP locations. To address limitations of current RNA-seq alignment and quantitation methods, we developed the complimentary tools Seqnature and EMASE. Seqnature incorporates known SNPs and indels into reference genomes, and constructs personalized genomes. EMASE employs an Expectation Maximization (EM) algorithm to estimate gene expression and ASE simultaneously by apportioning multi-reads at the level of gene, isoform, and allele. We show that accounting for individual genetic va