Protein Analysis

[tedrik]

---

Members do not see this ad.

Hi everyone,

My PI told me to try to find some sort of protein sequence analyzer (on the web) that allows you to type in a short sequence of amino acids and find out what other proteins are known to bind to or interact with it. I've been looking for a few days now and I thought I would just post this up and see if anyone has any experience with such a service.

Thanks a lot!

 

[Fixed Gear]

The closest thing I know of that fits the identity part is NCBI's BLAST.

You input a protein by primary AA sequence and it searches for identical/similar/homologous protein sequences. I cannot offer any advice on identifying interacting proteins. Maybe someone else can.

the BLAST interface is here:

http://www.ncbi.nlm.nih.gov/BLAST

Given that you are doing protein-protein homology, you'll want the 'Blastp' option.

 

[BDavis]

Hi everyone,

My PI told me to try to find some sort of protein sequence analyzer (on the web) that allows you to type in a short sequence of amino acids and find out what other proteins are known to bind to or interact with it. I've been looking for a few days now and I thought I would just post this up and see if anyone has any experience with such a service.

Thanks a lot!

I don't know what system you are particulary interested in, but for mammalian proteins I use the following sites (seems like a lot, but if you stare long enough something will catch your eye).

http://www.cbio.mskcc.org/prl/index.php

Good luck.

 

[tedrik]

I think these will be nice places to start. Thanks for the responses guys, I appreciate your suggestions!

 

[tr]

My PI told me to try to find some sort of protein sequence analyzer (on the web) that allows you to type in a short sequence of amino acids and find out what other proteins are known to bind to or interact with it.
Thanks a lot!

As far as I know this is impossible. Since right now we can't even predict the structure of a protein from its amino acid sequence, goodness knows how one might try to go about predicting its interacting partners a priori.

One thing you can do is (as other posters have suggested) try a homology search. If you identify highly conserved domains that already have known interacting partners, those partners will be reasonable candidates to interact with your protein as well. You can then go do a co-immunoprecipitation or GST pulldown to check for an interaction.

Alternatively, I know there is a Japanese group that has begun to compile an enormous database of protein-protein interactions. They do yeast two-hybrids on a massive scale (I think they've used something like 15,000 clones so far), and their results are freely available on the web. If you are lucky, your protein may be in there already. Check out

http://www.kazusa.or.jp/huge/ppi/