2
$\begingroup$

Lets say I only have a list of gene names. I know they are up regulated and I know they are related to cancer.

What information can I obtain from having only the names of genes? For example is there any way to find a gene-gene interaction, or anything else? Is it possible to retrieve any information by having only the name of the genes?

$\endgroup$
  • 1
    $\begingroup$ The info you have is the gene name, nothing more, nothing less. You need info about each gene to draw any conclusions. $\endgroup$ – MattDMo Apr 1 '15 at 17:04
  • $\begingroup$ @MattDMo thanks for your message. how can I get info about each gene , do you know any way to do it? should i search one by one in google and see what comes up ? $\endgroup$ – Learner Apr 1 '15 at 17:08
  • $\begingroup$ I don't really understand what you're asking. With a gene's name, you can find out anything about it that's ever been published $\endgroup$ – canadianer Apr 1 '15 at 18:02
  • $\begingroup$ @canadianer Yes when it comes to only one gene but when you have a list of them, how they interact with each other or anything else related to a set and not one gene alone $\endgroup$ – Learner Apr 1 '15 at 18:11
  • 2
    $\begingroup$ You can do a GSEA using 1 set, the other one coming from reference databases such as MSigDB (see here). You can categorize your list by gene families for example. $\endgroup$ – cagliari2005 Apr 1 '15 at 20:01
7
$\begingroup$

I would suggest searching the name in any trusted genetics database such as NCBI's GenBank (http://www.ncbi.nlm.nih.gov/genbank/). You can just Google search it, but it may take a little longer to find useful information that way.

I hope this helps and good luck with your research,

CDB

$\endgroup$
  • $\begingroup$ NCBI also hosts OMIM, which is a good place to find a link between a gene name and a disease occurring of the gene is mutated. You can select OMIM by following the link above and then selecting OMIM from the drop-down menu by the search box. Then search the gene name. $\endgroup$ – Jon D. Moulton Apr 8 '15 at 14:29
5
$\begingroup$

Was getting long in the comments.

In light of your comments, you might be interested in Gene-set enrichment analysis (GSEA). You can do a GSEA using your set, the other one coming from reference databases such as MSigDB (see here). You can categorize your list by gene families using this technique for example.

You can get an idea of what cellular process is the most impacted by the genes in your list and from there try to reach some conclusions but just as a warning, what you are attempting is not trivial at all. It will be hard to reach valid conclusions from a simple list of genes as correlation does not mean causation (i.e. is the gene up regulated because of cancer or is it provoking the cancer?)

To visualize pathways you might want to look at this post on Biostars. Several software are available for this purpose.

$\endgroup$
3
$\begingroup$

It depends; what species are the genes from? Some organisms have extensively annotated genomes, and actively curated species-specific databases, while other species may not even have a reference genome sequence available. By itself, a priori, if you were lucky, about all you list would tell you was how to spell the names of those genes--if you're lucky. But to a skilled Bioinformatician said list could be a gateway to a discovery, or a patent, or a publication, or a grant, although none of these is very likely.

$\endgroup$
  • $\begingroup$ I know already the genes are human genes $\endgroup$ – Learner Apr 1 '15 at 18:30
2
$\begingroup$

Also try putting the list through Reactome or String DB to see if you see mapping to certain pathways. http://string-db.org/

You can also put lists through ConceptGen to carry out ontology based analyses http://portal.ncibi.org/gateway/conceptgen.html

$\endgroup$
2
$\begingroup$

Assuming each gene symbol represents a unique gene or protein (this may not be true; see for gene name aliases in e.g. NCBI Entrez gene) you can get a lot of information programmatically. Below is an example using R and Bioconductor resources.

Define your list of genes:

# list of gene symbols, here we focus on one.
> genes <- "KRAS"

Load the annotation package org.Hs.eg.db and retrieve basic information including Gene Ontology ids:

# annotation package for human genes.
> library(org.Hs.eg.db)

# query Entrez gene id, description (GENENAME) and Gene Ontology id (GO)
> info <- select(org.Hs.eg.db, keys = genes, keytype = "SYMBOL", columns = c("ENTREZID", "GENENAME", "GO"))
> head(info)
  SYMBOL ENTREZID                                   GENENAME         GO EVIDENCE
1   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0000165      TAS
2   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0000186      TAS
3   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0001934      IMP
4   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005515      IPI
5   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005525      IEA
6   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005737      IDA
  ONTOLOGY
1       BP
2       BP
3       BP
4       MF
5       MF
6       CC

Load the GO.db package and retrieve the term associated with the GO ids (this gives info about putative function):

> library(GO.db)
> info$Term <- Term(GOTERM[info$GO])
> head(info)
  SYMBOL ENTREZID                                   GENENAME         GO EVIDENCE
1   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0000165      TAS
2   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0000186      TAS
3   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0001934      IMP
4   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005515      IPI
5   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005525      IEA
6   KRAS     3845 Kirsten rat sarcoma viral oncogene homolog GO:0005737      IDA
  ONTOLOGY                                           Term
1       BP                                   MAPK cascade
2       BP                   activation of MAPKK activity
3       BP positive regulation of protein phosphorylation
4       MF                                protein binding
5       MF                                    GTP binding
6       CC

Load the RefNet package and get information about interacting proteins:

> library(RefNet)
> refnet <- RefNet() # this step downloads from AnnotationHub and may take some time.
> int <- interactions(refnet, species = "9606", id = genes, provider = "BioGrid")
# create data.frame with Entrez id of proteins.
> d <- int[,1:2]
> d$A <- sub(".*locuslink:(.*)\\|.*", "\\1", d$A)
> d$B <- sub(".*locuslink:(.*)\\|.*", "\\1", d$B)
> head(d)
      A    B
1  3065 3845
2  8841 3845
3 23411 3845
4  1994 3845
5  3845 9770
6  9770 3845

Use the igraph package to visualize the interacting proteins.

> library(igraph)
> g <- igraph::simplify(graph.data.frame(d, directed = FALSE))

# annotate the nodes with the symbol.
> V(g)$label <- select(org.Hs.eg.db, keys = V(g)$name, columns = c("SYMBOL"), )$SYMBOL
> plot(g)

Interaction network of human KRAS

Then you could repeat the cycle with the interacting partners. There are many other things you can do, starting from a list of symbols; this is just a brief example. More detailed examples and workflows can be found here.

$\endgroup$
0
$\begingroup$

take the gene names and copy all of them, next open an online tool called gene mania. paste all your genes on the window. you will get their interaction and other related genes. Now go to a database called gene cards and type in all the gene you got from gene mania and paste it one after the other to get useful information about the individual genes. some of which are restriction site information's, transcription factors etc.

$\endgroup$
  • 1
    $\begingroup$ It would be helpful if you provided, at least, the URL of the tool you recommend. $\endgroup$ – ddiez May 22 '15 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.