MytiBase: a knowledgebase of mussel (M. galloprovincialis) transcribed sequences

Abstract

Background: Although bivalves are among the most studied marine organisms due to their ecological role, economic importance and use in pollution biomonitoring, very little information is available on the genome sequences of mussels. This study reports the functional analysis of a large-scale Expressed Sequence Tag (EST) sequencing from different tissues of Mytilus galloprovincialis (the Mediterranean mussel) challenged with toxic pollutants, temperature and potentially pathogenic bacteria.

Results: We have constructed and sequenced seventeen cDNA libraries from different Mediterranean mussel tissues: gills, digestive gland, foot, anterior and posterior adductor muscle, mantle and haemocytes. A total of 24,939 clones were sequenced from these libraries generating 18,788 high-quality ESTs which were assembled into 2,446 overlapping clusters and 4,666 singletons resulting in a total of 7,112 non-redundant sequences. In particular, a high-quality normalized cDNA library (Nor01) was constructed as determined by the high rate of gene discovery (65.6%). Bioinformatic screening of the non-redundant M. galloprovincialis sequences identified 159 microsatellite-containing ESTs. Clusters, consensuses, related similarities and gene ontology searches have been organized in a dedicated, searchable database http://mussel.cribi.unipd.it.

Conclusion: We defined the first species-specific catalogue of M. galloprovincialis ESTs including 7,112 unique transcribed sequences. Putative microsatellite markers were identified. This annotated catalogue represents a valuable platform for expression studies, marker validation and genetic linkage analysis for investigations in the biology of Mediterranean mussels.

Figure 1

Representation of length distribution of EST (A) and consensus sequences (B).

Figure 2

Snapshots of the MytiBase web interface. The search interface (A) allows users to filter data fields, (i.e. number of ESTS in a cluster, or words in cluster description). Following the cluster/EST link, a detailed report (B and C) can be accessed.

Figure 3

Organisms most represented in the protein similarity searches with mussel sequences. Percentages of transcripts finding Blast-X similarity (E-value < e^-6) in the non-redundant protein database attributable to a given organism (manually examined annotations). The organism names underlined are those with statistically significant (p-value = 0) representation in our MytiBase respect to the number of protein sequences in UniProtKB database.

Figure 4

Gene Ontology categorization of 3,275 M. galloprovincialis annotated sequences. The total numbers of consensus sequences classified in each main GO category are 1,767 for Biological Process, 2,266 for Molecular Function and 1,543 for Cellular component. Since a gene product could be assigned to more than one GO term, the percentages in each main category do not add up to 100%. See the Additional files 3, 4 and 5 for more details.