Lycopersicon Combined Unigene Build Series

This unigene build series incorporates ESTs derived from Lycopersicon hirsutum, Lycopersion pennellii, and Lycopersion esculentum cDNA libraries. No other sequences are incorporated at this time. These libraries were constructed at Cornell University as part of the NSF funded Tomato Genomics Project (#9872617), and sequenced predominantly by TIGR. In pre-funding stages of the project, pilot sequencing was also provided by Cereon and Novartis. All sequences are 5' reads, except approximately 1% of the clones, selected at random, were also sequenced from the 3' end.

Summary of new features in this build

  • New 5' and 3' sequences from re-arrayed "TUS" library
  • New trimming and quality evaluation process
  • New chimera screening processes reduce number of chimeric sequences introduced to the assembly process
  • rRNA and cloning host contamination screened out
  • BLAST results against NR and other databases cached and displayed automatically
New Sequences

New in the latest iteration of this build is the creation and incorporation of a "re-arrayed" library which contains clones selected from the set of plates originally sequenced, spanning all of our cDNA libraries. Clones were selected to span our previous Lycopersicon unigene build as well as all of the clones used on the publicly available Tomato cDNA microarray. Efforts are in progress to (re)sequence this set of clones from both 5' and 3' ends and incorporate these new, paired reads, into our unigene assemblies. While this additional sequencing project is not yet complete, the current Lycopersicon Combined build incorporates 11732 new 5' reads and 9897 new 3' reads.

Re-sequencing of TOM1 microarray clones was funded by INRA (French National Institute of Agronomics Research). For further information, contact:

Mondher Bouzayen (bouzayen@ensat.fr)
Genomics and Fruit Biotechnology Lab.
UMR 990 INRA/INP-ENSAT
Avenue de l'agrobiopole
BP107 Auzeville-Tolosan
F-31326 Castanet Tolosan Cedex, France

Additional funding for 5'/3' sequencing non-array TUS clones was provided by the Italian Ministry of Agriculture and Forestry (MiPAF) as part of project DM357/7303/01, and performed by Avesthagen Technologies Ltd., India. For further information, contact:

Chris Bowler (chris@szn.it)
Stazione Zoologica
Naples, Italy

New trimming and quality evaluation

Also new in this build is a completely redesigned raw data processing pipeline. All reads currently incorporated into SGN's unigenes are processed directly from the original chromatogram file. The high-quality portion of the cDNA insert is recovered by our own customized insert recovery process. [Details page under development]

Chimera Screening

Incorporated as well in this pipeline are 3 independent screens for chimeric sequences. While none of these screens have been validated yet in the laboratory, statistics collected during EST preclustering show substantial reduction in putative false joining of EST clusters. ESTs considered to be putative chimeras are censored from the assembly process, reducing the false representation of spurious cDNA ligations as novel genes in the unigene output. [Details page under development]

BLAST results stored

With the current Lycopersicon Combined build and future unigene builds, BLASTs against common databases such as the genbank non-redundant peptide database (genbank/nr) and the Arabidopsis predicted proteome (TAIR) are precomputed and stored in SGN's databases. Stored matches can be viewed on unigene search result pages as a simple first pass annotation, where matches exist. [see example]

Using this unigene build

Utilizing the unigene build is as simple as searching against it. The most straight-forward way to search is to use the SGN BLAST interface and select Lycopersicon Combined as the target database. Note that this BLAST database is nucleotide sequence, so you must use BLASTN, TBLASTN, or TBLASTX to search it. The resulting matches will provide links to detail pages for those unigenes.

Another way to use this unigene build is to search it directly with an SGN-U# identifier, or search the EST database with an EST identifier. The former requires you to have already identified the unigene before and noted its SGN-U#, or to have noted a reference somewhere using this identifier. The EST database however can be searched with genbank accessions, facility assigned identifiers, and clone stock identifiers, in addition to SGN's native SGN-E# identifiers.

Finally, uses of TIGR's tomato gene index may search against this and other Lycopersicon builds using TIGR's TC#s. Any current TIGR TC# or older numbers from previous releases for which TIGR still maintains tracking information can be used to identify SGN unigenes which are assemblies of sequences in common those assembled in TIGR's tentative consensus assembly. Note that this mapping is not one-to-one, but many-to-many. Furthermore, our build contains nearly 20,000 sequences not contained in TIGR's most recent gene index release for tomato, and their build contains input sequences from exogenous sources which we did not include in this build.