Bioinformatics Stack Exchange is a question and answer site for researchers, developers, students, teachers, and end users interested in bioinformatics. It only takes a minute to sign up.
Sign up to join this community
For proteins that have multiple chains (e.g., 1EMS), is there an easy way to specify which chain I want to use for blastp?
I cannot imagine that I am the first person to have this problem, but so far my only solution has been to write my own script that gets the chain I want from the .pdb file and convert that to a .fasta file.
The NCBI BLAST can be used with the PDB DB (which NCBI has). The PDB codes are stored as 4 letter codes underscore chain, e.g. 1GFL_B. The catch is segment identifiers —but generally they are the same peptide so shouldn't be an issue.
You can search specifically the PDB DB in NCBI (not the RCSB PDB) by setting the database to PDB.
_B in the query (ie. the thing you are search with) is the chain of the PDB structure., whereas the 4-letter code is the PDB structure as appears in the RCSB PDB or PDBe. Namely, NCBI does not store the complex as a single file
$\endgroup$
Nov 25, 2020 at 16:08
standalone, within the NCBI, within Uniprot and (badly) within RCSB PDB. So for getting help being specific with questions helps or else you get unhelpful answers —like mine— or worse get your question closed due to apparent lack of research.
$\endgroup$
Nov 25, 2020 at 16:19
WEB VERSION OF BLASTP
When you run a blastp query using a fasta file with multiple chains you can choose which chain's results to examine using a drop-down box that appears after the search has run.
TERMINAL BLASTP
The results for each chain will be listed separately. For ease of searching you can specify the output to be a csv of your desired format (See blastp -help and look up the -outfmt option for more information).
Note that for some proteins, such as the 1EMS you chose for an example, the multiple chains are identical. When you get the fasta file for that specific protein it will only have one entry. However if you really need a fasta file formated so each chain has its own sequence you can get the fasta file for all PDB entries and extract the specific protein_chain you want.