Identification and analysis of common bean (Phaseolus vulgaris L.) transcriptomes by massively parallel pyrosequencing

Abstract

BackgroundCommon bean (Phaseolus vulgaris) is the most important food legume in the world. Although this crop is very important to both the developed and developing world as a means of dietary protein supply, resources available in common bean are limited. Global transcriptome analysis is important to better understand gene expression, genetic variation, and gene structure annotation in addition to other important features. However, the number and description of common bean sequences are very limited, which greatly inhibits genome and transcriptome research. Here we used 454 pyrosequencing to obtain a substantial transcriptome dataset for common bean.ResultsWe obtained 1,692,972 reads with an average read length of 207 nucleotides (nt). These reads were assembled into 59,295 unigenes including 39,572 contigs and 19,723 singletons, in addition to 35,328 singletons less than 100 bp. Comparing the unigenes to common bean ESTs deposited in GenBank, we found that 53.40% or 31,664 of these unigenes had no matches to this dataset and can be considered as new common bean transcripts. Functional annotation of the unigenes carried out by Gene Ontology assignments from hits to Arabidopsis and soybean indicated coverage of a broad range of GO categories. The common bean unigenes were also compared to the bean bacterial artificial chromosome (BAC) end sequences, and a total of 21% of the unigenes (12,724) including 9,199 contigs and 3,256 singletons match to the 8,823 BAC-end sequences. In addition, a large number of simple sequence repeats (SSRs) and transcription factors were also identified in this study.ConclusionsThis work provides the first large scale identification of the common bean transcriptome derived by 454 pyrosequencing. This research has resulted in a 150% increase in the number of Phaseolus vulgaris ESTs. The dataset obtained through this analysis will provide a platform for functional genomics in common bean and related legumes and will aid in the development of molecular markers that can be used for tagging genes of interest. Additionally, these sequences will provide a means for better annotation of the on-going common bean whole genome sequencing.

Recent transcriptomics advances and emerging applications in food science

“…The reads delivered by the sequencer are filtered and mapped to a reference genome (or transcriptome), to be subsequently counted and normalized with respect to transcript length and sequencing depth. The FLX platform in combination with barcode tagging was also demonstrated to be a good option for quantitative gene-expression analysis for the common bean [59]. Tagging cDNA libraries was very helpful in verifying and validating global gene-expression patterns, and detecting both shared and unique transcripts among the bean tissues analyzed.…”

Genomics of Origin, Domestication and Evolution of Phaseolus vulgaris

“…Next-generation sequencing (Mardis 2008) has revolutionized the "-omic era", allowing the analysis of millions of reads in a very little time and at much reduced cost. Kalavacharla et al (2011) provided new genomic information by sequencing a large number of cDNA libraries from different plant tissues using the Roche 454-FLX pyrosequencing platform: leaves, flowers and roots from a common bean cultivar, and pods derived from the BAT93 breeding line, one of the parents of the core common bean mapping populations. They identified 59,295 common bean unigenes, 31,664 of which were newly discovered sequences.…”

Marker-Assisted Molecular Profiling, Deletion Mutant Analysis, and RNA-Seq Reveal a Disease Resistance Cluster Associated with Uromyces appendiculatus Infection in Common Bean Phaseolus vulgaris L.

“…We analyzed transcriptomes for two biological replicates of “Sierra” and crg, including 12 h post inoculation (HPI) MI and I samples for each of the two replicates, totaling eight libraries of 175 million reads (Table 2). In common bean [10], soybean ( Glycine max ) [19], and related studies [21], hypersensitive response (HR) has been shown to be expressive between 0–12 hpi. Therefore, 12 hpi is the time-point chosen for our study [22].…”