Improved Assembly of the Asian Seabass Transcriptome towards obtaining Full-Length Coding Sequences

The Asian seabass (or barramundi; Lates calcarifer) is a carnivorous marine teleost and a popular food fish in Southeast Asia. It has an estimated genome size of ~700Mb which has not been sequenced yet, while its transcriptome is poorly represented by just over 22,000 EST sequences in GenBank. Previously, we obtained mRNA sequence data for samples pooled from various organs of multiple individuals using three different platforms, namely SOLiD 3+, 454 FLX Titanium and paired-end Illumina HiSeq 2000. We performed a co-assembly of all these data using a combination of Trinity, CLCBio and CAP3, by which we were able to extract 6,287 full-length coding sequences. Here, we report an improved approach to the assembly procedure in order to give greater emphasis to extracting full-length coding sequences. In the current approach, we focused on assembling only the HiSeq data first, by splitting the data files into smaller sets, performing an assembly of each set, and extracting the full-length sequences. The unused reads from all these sets were then collected, and a second round of assembly was performed in similar manner. This approach has resulted in an increase to 11,418 full-length coding sequences from the first two rounds of assembly, corresponding to ~5000 unique Nile tilapia full-length protein sequences. The work is currently in progress to assemble the remaining unused reads, potentially by including the data from the other sequencing platforms as well. This assembled transcriptome will provide an essential foundation for development of tools for gene expression analysis and genome annotation.