I have used panther data base to detect the pathways for set of genes , I entered the gene symbol ( 500) gene but id does not work ? the data set is an Agilent.
We need you to clarify your question. What do you mean by using "Panther database to detect pathway genes" ? What tool are you using ? Where did you enter gene symbols ? What do you mean by "it doesn't work" ? Do you get an error message ?
Also make sure you're using correct gene symbols (with respect to the species of interest).
@jean I used the link bellow:
and in the text box which is under:
Enter ids and or select file for batch upload. Else enter ids or select file or list from workspace for comparing to a reference list:
I entered the genes symbol:
then when I select the enrich analyse and panther pathway, I got empty result? but when I used reactom I have result?
The two most likely explanations are that either you're not using symbols/identifiers that Panther recognizes or that Panther doesn't have pathways with your genes of interest. Different pathway databases have different notions of pathways and different ways of curating them and none of them is comprehensive in terms of gene coverage. When doing enrichment analysis, it's also possible that your list is not enriched in any pathway represented in Panther. You may get enrichment in a pathway with the same or similar name in another database e.g. Reactome due to the differences in pathways compositions.
@Jean , The Panther DataBase give some result if I remove the check box for :bonferroni correction for multiple test??
It means that there's no statistically significant result at the selected threshold after applying the Bonferroni correction for multiple testing. The problem is that the Bonferroni correction is too conservative for most practical cases and GO terms enrichment in particular.
@jean: I have a significant pathways from Reactome and significant biological process from GO (no pathways from panther data base), I am confused , :( please I need your help :(
I don't know the details of what you've tried but the main reason you get different results is simply because different tools/databases do things differently. First, as I mentioned previously, the different databases have different pathways (i.e. they group genes differently), second the associated enrichment tools most likely also use different background gene lists, third they may also implement different testing procedures (like for example how to deal with redundancy/overlap in particular for GO terms) with different corrections for multiple testing (or maybe none at all). Note, on the matter of background gene lists, that a proper background list should be composed only of the genes interrogated in the experiment used to derive the query list. For genome-wide studies, using the whole genome as background is fine but for example, if you studied only genes on one chromosome, then your background list should be composed only of genes from that chromosome.
You may want to read more about pathway analysis in other posts on biostars or in this review.
Login before adding your answer.
Use of this site constitutes acceptance of our User Agreement and Privacy