BMC Genomics. 2008; 9: 508.
Joan Cerdà et al.,
As a former expert of EST technology and analysis, I still really enjoy reading the state-of-the-nation in EST papers, and like to see how the technological envelope is being opened further. While I suspect that rather many research groups are under-selling their EST collections, and are completely failing to fully exploit their own data, this manuscript manages to add something new to the EST genre.
The Senegalese sole is a flatfish of economic relevance within Europe and North Africa. The fish is within an aquaculture development programme, but physiological aspects of growth and development including at least disease resistance and larval growth remain uncontrolled leaving room for substantial improvements. This manuscript concentrates on the development of genomic resources for the study of gonad development in the fish, and within a systems-scale analysis the authors include sequence data from cDNA libraries, and a substantial amount of in situ data and this is wrapped into what appears to be a very attractive data presentation environment.
10 high titre cDNA libraries were constructed from different developmental stages, tissues and organs, and 3′ sequencing was used to obtain a total of 5,200 EST sequences. The sequences were processed with a rather primitive bioinformatics analysis pipeline, but a meaningful unigene set was assembled and cohorts of meaningful tentative consensus sequences were identified. The core analyses were based solely (pun intended 
) on metrics such as number of ESTs represented within unigene and GO mapping of unigenes on basis of BLAST results. This certainly yields the standard but appealing eye-candy and demonstrates a grasp of the data (but value beyond the aesthetic is questionable).
The sequences were used to create an Agilent custom expression array, and this has been demonstrated to work, although lists of differentially expressed gene expression within meaningful comparisons are likely to follow in subsequent manuscripts.
The core value of this manuscript is however their Soleamold bioinformatics application – needs to be installed on Windows (why couldn’t they have packaged a Java Webstart application instead) – but shows a great set of screen shots of morphology and ISH data whereby the genomic, transcriptomic and ISH data are integrated into a single coherent application.
Overall, this is a great manuscript demonstrating what can be done with just a few 1000s of EST sequences, flexible technologies such as the Agilent custom arrays and a load of IST. The bioinformatics of data analysis is poor and incomplete, but the integrative imagination and implementation looks five-star. Great read, good concept and nice implementation. This is undoubtedly worth a read-of-the-week!