Hi, not exactly. I tried to clarify in the question. What I would expect your command line to return is basically 2r9r: A, B. Interestingly it only returns 2r9r_B, maybe I don't completely understand it. Anyway what I need is for every chain id which other chain ids it is equal to. I hope that makes it clearer?
List Of Sequence Unique Pdb Chains
Hi everyone,
I need a list of which PDB chains are sequence-unique.
For example. In PDB entry 2r9r there are 4 chains. But some have equal SEQRES sequences. See here. It says:
The structure 2R9R has in total 4 chains. These are represented by 2 sequence-unique entities.
The chains are A,B,G,H where A is equal to G and B equal to H. I want the following sets:
2r9r: (A,G), (B,H)
The naive approach would be downloading the FASTAs and doing the check myself. This file might be a suitable source:
ftp://ftp.wwpdb.org/pub/pdb/deriveddata/pdbseqres.txt
I'm wondering though, whether such a list does not exist already? I'm also open for suggestions on how to better built it oneself.
*Edit: To clarify again: I can compile this myself, the main point is more whether the list is not available already. That would be more convenient than iterating the full sequence data every time something changes.
Thanks,
Jonas
4 answers
I'm not sure I understand your problem and what is the link to the file. Do you just need the following command line ?:
curl -s "ftp://ftp.wwpdb.org/pub/pdb/derived_data/pdb_seqres.txt" |\
tr "\n>" "\t\n" |\
sort -t ' ' -k2,2 -u
• 0 views
•
link
written
by
Pierre Lindenbaum
You're right it's no big deal. I just would have preferred it to downloading the full set of FASTA sequences in the PDB every time. As they give that information on their website it seemed reasonable to me that it is stored somewhere, but that does not seem to be the case.
Have you tried ASTRAL - SCOP Original style datasets ?
• 0 views
•
link
written
by
Khader Shameer
See the NCBI Non-redundant PDB chain set.
A summary data file is available from the FTP site. Note: check the dates of the files; the latest is called nrpdb.090711.
The file is plain text with 26 columns. Entries for your PDB example look like:
# 1 2 3 4 5 6 7 8 9 A B C D E F G H I J K L M N O P Q
2R9R B 60830 167 43 0 446 29 0 474 3 0 30897 1 1 0.00 27.82 24.90 3.89 2.40 4 27 3 514 X a
2R9R H 60830 167 44 0 446 30 0 474 4 0 30897 2 0 0.00 29.97 29.38 0.83 2.40 4 27 3 514 X a
2R9R A 60830 200 6 0 2531 4 0 2379 4 0 21166 1 1 0.00 2.10 2.10 0.00 2.40 4 27 3 333 X a
2R9R G 60830 200 7 0 2531 5 0 2379 5 0 21166 2 0 0.00 2.10 2.10 0.00 2.40 4 27 3 333 X a
where columns 4, 7, A and D are Group IDs (at various clustering levels), showing that chains (B, H) and (A, G) are grouped.
However, as others have stated, it really is very quick and easy to generate this kind of data from a file of PDB chain sequences and a tool such as CD-HIT.
Log in to answer this question.
More posts like this
-
Find if two proteins have contacts - Python
written by albert.castella.teruel •
-
Expected number of read duplicates
written by dariober
-
Calculating Gc Content For All Ccds
written by Davy
-
Indexing A List For Extracting X,Y,Z Coordinates From A Pdb File
written by s.charonis •
-
Statistics Of Enrichment Of Indels
written by PoGibas
-
Interpro Parentchildtreefile -> .Obo File
written by boczniak767
-
What Is A Good Way To Implement Getting A Consensus Sequence In Java?
written by Brandstaetter
-
Which Bioperl Module Can I Use For Parsing/Mining Blast Output?
written by Raghul
-
How To Judge If A Pfam Domain Is A Repeat And The Functional Similarities Of A Series Domains?
written by Ct586
-
Parsing Dimplot Result Files And Annotation
written by Samimirza •
