Comprehensive Database Searching

The protein matches and the InterPro analysis have already given functional clues about our novel protein. However if this particular gene product was located in close proximity to an SNP with a disease association we would need to find out as much as possible, not only to provide more supporting evidence for the gene product but also testable predictions about function that can be followed up. Performing a comprehensive search is not a trivial exercise since it involves 17 divisions of GenBank and sources of trace data that have not yet been submitted to GenBank. So where do we start? The two large repositories labelled nr protein or nucleotide on the NCBI BLAST server are a useful first choice ( We have already checked nr protein at 891,607 sequences but we need to compliment this with month, which in this case yields another 61,254 protein sequences but no additional high-scoring hits. The search against nr nucleotide with 1,192,858 sequences records three extended matches. This includes the mouse sequence already described, BC023073, and the primary accession number of the finished genomic section AP000892. The third match, XM_100696, is a secondary accession number for a reference mRNA sequence predicted by the NCBI Annotation Project from a genomic contig NT_009151. This is the same prediction labelled LOC160162 in Figure 4.5. There is an accompanying 56-residue predicted ORF that is in the NCBI protein database but has no supporting evidence. Inspection of the genomic location suggests it may be a spurious prediction.

Checking public patented proteins at 88,019 sequences gave no hits. However the patent nucleotide division, gbPAT, at 581,001 sequences, gives three solid hits, AX321627, AX192589 and AX072029. The first of these is a 2114-bp DNA from patent WO0172295. The document indicates this protein was isolated from a lung cancer sample (http://ep. These hits constitute a partial mRNA level of confirmation for the novel protein but a reciprocal check (i.e. a BLASTN of AX321627 against the nr nucleotide database) indicates this clone may be a chimera from two separate gene products. A search against a commercial patent database, containing 673,453 protein sequences, reveals identity matches for the N-terminal section from patent WO200060077 and a C-terminal identity match from WO200055350, both of which are reported as cancer-associated transcripts ( Checking the GSS division by TBLASTN gives four genomic hits; AZ847251 from mouse, AG114530 from chimpanzee, BH306228 from rat and BH406519 from chicken. Using BLAST against the

TABLE 4.1 Useful Resources for Gene Finding and Analysis

Site description

Ensembl at EBI/Sanger

Centre Human Genome Browser at UCSC Map Viewer at NCBI Protein Atlas of the genome SWISS-2DPAGE

database Ensembl 4.28.1

announcement NCBI gene model builder UniGene EST clusters InterPro at EBI Proteome analysis at EBI Google general search portal RefSeq at NCBI International Protein Index

Derwent sequence patent databases BLAST at NCBI BLAT at UCSC DAS—distributed annotation Exofish at Genoscope Fgenesh at Sanger

Institute Expasy translation tool CAP3 nucleotide assembly tool GeneWise at Sanger

Institute Genscan at MIT SSAHA at Sanger Institute search

ModelMakerHelp.html = start)

Ensembl mouse peptides detected a C-terminal similarity that is a zinc finger domain match. However both the human and mouse mRNA have unique and solid hits against mouse chromosome 9.40 Mb. This suggests the gene product is derived from this locus although it has not been annotated yet by Ensembl. Interestingly the gene lies between two odour receptors, unlike the human positioning between BACE and PCSK7, showing the position is non-syntenic.

Drawing detailed conclusions from these results is outside the scope of this chapter but the example makes clear how much extra information a comprehensive database search can yield. Was the protein unknown and/or novel? The difficulty of answering this question illustrates the diminishing utility of these terms. The protein has at least one function-related motif that can be recognized at high specificity so it can no longer be classified as an unknown. It remains novel only in the strict sense of not being represented in the current protein databases. It is not novel in the wider sense because both the mRNA and ORF were substantially covered as predicted by sequence data entries in the public and patent databases respectively.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment