Dear all,
Your help is needed here: We have a list of 700 proteins: is it possible to find out, which ones are relevant for pain and inflammation without having to check them manually?
Something like a search engine to find out in which pathways are these proteins involved?
Thanks in advance!
Best, Alaa
Hi Alaa,
To do it avoiding any manual checking, you should have a little programming knowledge. I have done similar things before, so I suggest you the following steps:
- KEGG is a large database whose contents range from genes to pathways, so here you may find what you are looking for, provided that you know the pathways that are involved in "pain and inflammation". To retrieve bulk data directly from the database, there is the KEGG REST API. Here you will find the basics about how to use it (http://www.kegg.jp/kegg/rest/keggapi.html). So, my first step will be retrieving the pathways for each protein ID you have in your set. But you will need some programming skills to accomplish this task.
-Perl allows you a flexible handling of biological databases, so I recommend you using it. Installing a module called LWP::Simple, will let you to use the function 'get'. This, combined with the KEGG REST API, is useful for storing KEGG data into variables, and processing them. So, if you start a new script, and you type '$kegg_data = get("http://rest.kegg.jp/get/hsa:$geneID")', the variable kegg_data will contain all the info related to a certain gene ID. If you do this for every gene ID in your set, and search for desired KEGG pathway ID/s (the ones involved in pain and inflammation), you will be able to determine which genes are involved in pain and inflammation pathways.
-As you are starting from a protein set, I guess you won't have any gene ID. Using Dbfetch (http://www.ebi.ac.uk/Tools/dbfetch/) combined with Perl will allow you to find all the cross references related to a certain protein ID. Among those cross references you should be able to find any gene ID/s related to your protein ID. Use this gene IDs set as starting point to accomplish the previous steps.
-As a summary: Protein ID --> Get Gene ID (use dbfetch and Perl) --> Get pathway ID (use KEGG REST API and Perl) --> Discard Gene IDs not involved in desired pathways (Perl and regular expressions) --> Go back to protein ID --> You have protein IDs related to desired pathways
You may either find this info confusing, or bad explained, or simply too easy to waste so much time explaining every step. As I don't know how keen on programming are you, I tried to be as explicit as possible. Please notice me if you don't get my point, and I'll try to detail everything more, or writing a script for you (whichever you think is faster).
That was pretty informatie!! Will check and get back to you. Thank you very much!!!
Hi Alaa,
One more addition to the already extensive information provided by Juan above:
You can use STRING at:
http://string-db.org/
It is very easy to use. After pasting your protein list ((one per line), it takes you to pages that are enriched with pathways that are associated with your proteins.
Best,
Hediye
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Proteins