Integrating Transcriptomics and Proteomics: A Powerful Approach for the Identification of Protein Sequence Variants in the Genus Populus

Although traditional genome-wide association studies (GWAS) (i.e., SNP genotyping and trait measurements) are highly informative, they are not completely informative because the biological implications of most nucleotide alterations remain unverified. As such, GWAS studies require additional analyses such as gene expression analyses. With the availability of next-generation RNA sequencing (RNA-Seq), whole-transcriptome profiling is capable of obtaining wholesale analyses of genes and splices variants. Notably, the transcribed portion of the genome suggests which genes have been expressed; however, the identified transcripts may not be translated. Consequently, expression studies become more informative when coupled with proteome information. Thus, there is a need for bioinformatic methods that define and capture sample-specific transcripts, which can be used for mass spectrometry (MS) database searching. In principal then, RNA-Seq data could catalyze a paradigm shift in MS-based proteomic studies, providing an enabling platform for GWAS. Because of the high SNP frequency, using the Populus reference genome as a database proxy for closely related species has an increased risk of missing observable proteins. To overcome this limitation, we created sample-specific databases by aligning RNA-Seq reads to the P. trichocarpa reference genome. Using strict criteria, only high-quality sequence variants were retained. Unlike traditional searches, this approach provided the opportunity to profile an organism’s unique genetic structure (genotype) at the protein-level, which will inevitably provide a better understanding of how phenotypes emerge. Overall, 1000 expressed genetic variants have been characterized for two Populus genotypes. Interestingly, approximately 50 single amino acid polymorphisms were predicted to have a non-neutral effect.