What Online Gene-Centric Resources Do You Use?

Question

[Andrew Su]

You've done your microarray experiment (or your next-gen sequencing, proteomics, etc.) and you now have your list of candidate genes. What online gene-centric databases do you consult next to prioritize those candidates?

I'm interested in the big hitters in gene annotation like NCBI Entrez Gene, Ensembl, Uniprot, etc, as well as specialized databases that might be particular to your area (e.g., T1DBase for type 1 diabetes data, 4DXpress for expression data, OMA browser for orthology information, etc.)

I'm even interested in gene-centric reagent providers (e.g., Qiagen for siRNAs, Open Bio for clones, Santa Cruz Biotech for antibodies, etc.)

[Pierre Lindenbaum]

Wikipedia of course ;-)

[Andrew Su]

Pierre, if you don't put it as an answer, I can't up-vote and accept it as the best answer, which clearly it is... ;) (http://nar.oxfordjournals.org/cgi/content/full/gkp760v1, for the uninitiated...)

[Andrew Su]

Well, I was partially right. Just noticed that I can at least up-vote your comment...

Answer 1

[Sophia]

Here is another one that I find quite useful for a quick overview of in which processes a gene is implicated: Gene2Mesh from the NCBI

Search by gene ID and retrieve MeSH terms (= Medical Subject Headings). This resource is publication-based, and results are scored by significance of the association of the gene ID with the MeSH term

Answer 2

[Khader Shameer]

Along with Uniprot, ensembl and KEGG mentioned here, I used the following tools

• GSEA analysis is my first step
• OMIM to check if there is any known disease association
• KEGG pathway and diseases to check pathway related information
• ModBase to see if there is any scope for structural bioinformatics analysis
• STRING to get quick list of interacting partners, I usually check whether first degree partners have any role in any interesting pathways or diseases.

These days I am searching first in BioGPS using a custom layout to get a quick view from various databases (and saving a lot of time :))

[Andrew Su]

I must "like" any answer that mentions BioGPS. And in full disclosure, I asked this question to discover any other resources that can be added to the BioGPS plugin library...

[Giovanni M Dall'Olio]

I don't trust string at all, too many false positives/negatives.

[Khader Shameer]

You may replace STRING with any other PPI resources, but remember all of them have a good deal of FP or FN. And you need to literally grab PPI for your gene of interest from different database to get the coverage of PPI as in STRING. I prefer STRING because they apply a common scoring for such interactions from different resources, so that I can filter the PPI's am interested in.

[Khader Shameer]

From a users perspective, BioGPS is currently exceptional in its design and usability. Far better than Harvester or any other previous tools that tried to integrate multiple bioinformatics resources. It will be excellent if BioGPS can provide a platform for various analysis.

[Andrew Su]

Khader, this probably isn't the best forum for discussion (perhaps we can take it to email?), but I'm very interested to hear what you mean by "a platform for various analysis". (... and fyi, a huge performance enhancement is coming soon! Just walked out of our lead developer's office, and I'm amazed!)

[Khader Shameer]

Andrew, Yes we could discuss it over email. Good to hear that amazing enhancements are coming to BioGPS.

Answer 3

[Fred Fleche]

I am using KEGG in order to see which Pathway(s) have the more genes involved from the list of candidates genes.

So it gives me a listing of Kegg pathways ordered by the number of genes involved in the descendent order.

I am also a big fan of the Catalogue of Somatic Mutations in Cancer - COSMIC which provides mutation information for each gene in Cancer cell lines.

[Giovanni M Dall'Olio]

check reactome and its Sky Panter feature

Answer 4

[Giovanni M Dall'Olio]

I like very much Uniprot, it provides the best annotations for proteins: so, even if you are studying genes, I would have a look at that, to see the functions of the proteins coded.

For orthology, I use ensembl but I am aware that it is not the best resource for that. In fact, they predict orthology using only the longest CDS for each gene, so if you really care for orthology predictions you should look somewhere else.

And there is a wonderful tool called harvester, which allows you to obtain informations from many different databases at a time, for a certain gene: it cluster data from string, pubmed, kegg, etc...

Also, another gene-centric resource is GeneCards, which I think could be improved more but it is a sort of Uniprot for genes.

Answer 5

[Gww]

I like using the Reactome FI cytoscape plugin to look for interesting network modules.

[Andrew Su]

not an online resource per se, but looks interesting...

[Gww]

Reactome is online :). I just like to use the plugin.

Answer 6

[Thaman]

These are my top Gene tools

• Entrez Gene - NCBI's repository for gene-specific information
• Uniprot - Central repository of protein sequence and function even if we are studying specific gene.
• Ensembl - For wide varitey of gene annotation
• KEGG - Pathway related information
• OMIM - a database of human genes and genetic disorders
• Genecards - Database of human genes that provides concise genomic related information, on all known and predicted human genes. Moreover, GeneDecks, GeneTide, GeneAnnot, GeneLoc can be useful too.
• HGMD - The Human Gene Mutation Database
• GENSCAN - predicting the locations and exon-intron structures of genes in genomic sequences from a variety of organisms
• genprediction.org - All software related to geneprediction can be found here.
• genatlas.org -Contains relevant information with respect to gene mapping and genetic diseases

Regarding gene expression you can find several databases in HSLS

In R/Biconductor I have used topGO, GOstats, KEGGgraph and KEGGSOAP for enrichment and pathway analysis.

You can further check DAVID.

Answer 7

[Mary]

Huh--nobody mentioned DAVID. I really like DAVID for lists of stuff. But I also use most of the others that people mentioned as well.

[Thaman]

I mentioned it ;)

Answer 8

[Qdjm]

I might be biased but I think that GeneMANIA is a great tool for analyzing candidate gene lists.