Pfam

From Christoph's Personal Wiki
Revision as of 02:28, 6 September 2007 by Christoph (Talk | contribs) (See also)

Jump to: navigation, search

The Pfam database contains information about protein domains and families. Pfam-A is the manually curated portion of the database that contains over 8,000 entries. For each entry a protein sequence alignment and a hidden Markov model is stored. These hidden Markov models can be used to search sequence databases with the HMMer package written by Sean Eddy. Because the entries in Pfam-A do not cover all known proteins an automatically generated supplement is provided called Pfam-B. Pfam-B is derived from the PRODOM database.

The database iPfam builds on the domain description of Pfam. It investigates if different proteins described together in the protein structure database PDB are close enough to potentially interact.

Pfam MySQL database documentation

Note: This is taken directly from the Pfam website. It is archived here in case they take it offline.

Installation packages & documentation on MySQL can be obtained from www.mysql.com

Database structure

The table structures are in the '.sql' files found in .sql files.

Table data

The table data is in the form: "THE_TABLE_NAME".sql.gz (i.e., pfamA table data is in pfamA.sql.gz) The files can be downloaded from ftp://ftp.sanger.ac.uk/pub/databases/Pfam/database_files/

  • Upload files
  • gunzip files
  • Load into MySQL using:
mysql> load data infile 'FULL_PATH_DIR/pfamA.sql' into table pfamA;

Table queries

Query domains by architecture and protein ID/ACC

  • View proteins by Architecture

QUERY: Give me all the architectures and protein sequences for B12-binding family.

The architecture information is in 3 tables:

architecture
pfamA_architecture
pfamseq_architecture
select architecture, pfamseq_id 
from pfamA,architecture, pfamA_architecture, pfamseq_architecture, pfamseq 
where pfamA_id = "B12-binding" and pfamA.auto_pfamA = pfamA_architecture.auto_pfamA 
      and pfamA_architecture.auto_architecture = architecture.auto_architecture and 
      architecture.auto_architecture = pfamseq_architecture.auto_architecture 
       and pfamseq_architecture.auto_pfamseq = pfamseq.auto_pfamseq ;
  • View proteins by protein id/accession

QUERY: Give me all the domains for protein PLCG1_BOVIN

SEED ALIGNMENT:

mysql> select pfamA_acc, pfamA_id, seq_start, seq_end 
       from pfamseq, pfamA, pfamA_reg_seed
       where pfamseq_id = "pig1_bovin" and pfamseq.auto_pfamseq = pfamA_reg_seed.auto_pfamseq 
       and pfamA_reg_seed.auto_pfamA  = pfamA.auto_pfamA ;
+-----------+----------+-----------+---------+
| pfamA_acc | pfamA_id | seq_start | seq_end |
+-----------+----------+-----------+---------+
| PF00168   | C2       |      1090 |    1177 |
| PF00388   | PI-PLC-X |       321 |     465 |
| PF00018   | SH3_1    |       794 |     849 |
+-----------+----------+-----------+---------+
3 rows in set (0.03 sec)

FULL ALIGNMENT: This table is different from pfamA_reg_seed as it has the "in_full" column. You have to set the in_full = "1" if you want ONLY the domains that are in the full alignment and are significant. Otherwise it will include the insignificant matches and return thousands of rows!.

mysql> select pfamA_acc, pfamA_id, seq_start, seq_end 
       from pfamseq, pfamA, pfamA_reg_full 
       where pfamseq_id = "PLCG1_BOVIN" and pfamseq.auto_pfamseq = pfamA_reg_full.auto_pfamseq 
       and pfamA_reg_full.auto_pfamA  = pfamA.auto_pfamA  and in_full = "1";
+-----------+----------+-----------+---------+
| pfamA_acc | pfamA_id | seq_start | seq_end |
+-----------+----------+-----------+---------+
| PF00168   | C2       |      1090 |    1177 |
| PF00017   | SH2      |       550 |     639 |
| PF00017   | SH2      |       668 |     741 |
| PF00388   | PI-PLC-X |       321 |     465 |
| PF00018   | SH3_1    |       794 |     849 |
| PF00169   | PH       |        33 |     142 |
| PF00387   | PI-PLC-Y |       952 |    1070 |
+-----------+----------+-----------+---------+
7 rows in set (0.00 sec)
  • View other regions by protein id/accession
pfam-B regions
select distinct seq_start, seq_end, pfamB.pfamB_acc, pfamB_id
from pfamB_reg, pfamB, pfamseq 
where pfamseq_id = 'PLCG1_BOVIN' and pfamB_reg.auto_pfamseq = pfamseq.auto_pfamseq and pfamB_reg.auto_pfamB = pfamB.auto_pfamB
  • Other regions - transmembrane, signal-peptide, coiled-coils & low-complexity
select seq_start, seq_end, type_id, source_id, score
from other_reg, pfamseq
where pfamseq.pfamseq_id = 'PLCG1_BOVIN' and other_reg.auto_pfamseq = pfamseq.auto_pfamseq
  • Context regions
select seq_start, seq_end, domain_score, pfamA.pfamA_acc, pfamA_id,  pfamA.description
from context_pfam_regions, pfamseq, pfamA 
where pfamseq.pfamseq_id = 'PLCG1_BOVIN' and context_pfam_regions.auto_pfamseq = pfamseq.auto_pfamseq 
 and pfamA.auto_pfamA = context_pfam_regions.auto_pfamA
  • Smart regions
select seq_start, seq_end, smart_id;
from smart_regions, pfamseq, smart 
where pfamseq.pfamseq_id = 'PLCG1_BOVIN' and smart_regions.auto_pfamseq = pfamseq.auto_pfamseq 
 and smart.auto_smart = smart_regions.auto_smart 

Query domain by family ID/ACC

  • All domains for a family (FULL alignment)
mysql> select pfamseq_id, seq_start, seq_end, pfamA_id 
       from pfamA, pfamseq, pfamA_reg_full 
       where pfamA_id = "B12D" and pfamA.auto_pfamA = pfamA_reg_full.auto_pfamA
       and pfamA_reg_full.auto_pfamseq = pfamseq.auto_pfamseq and in_full = "1";
+------------+-----------+---------+----------+
| pfamseq_id | seq_start | seq_end | pfamA_id |
+------------+-----------+---------+----------+
| Q42338     |         2 |      88 | B12D     |
| Q9LJ47     |         1 |      87 | B12D     |
| Q9XHD5     |         3 |      89 | B12D     |
| O22414     |         3 |      89 | B12D     |
| Q940E1     |        29 |     116 | B12D     |
| Q40019     |         2 |      87 | B12D     |
| Q84MX3     |        87 |     173 | B12D     |
+------------+-----------+---------+----------+
7 rows in set (0.02 sec)
  • All domains for a family (SEED alignment)
mysql> select pfamseq_id, seq_start, seq_end, pfamA_id 
       from pfamA, pfamseq, pfamA_reg_seed 
       where pfamA_id = "B12D" and pfamA.auto_pfamA = pfamA_reg_seed.auto_pfamA
       and pfamA_reg_seed.auto_pfamseq = pfamseq.auto_pfamseq;
+------------+-----------+---------+----------+
| pfamseq_id | seq_start | seq_end | pfamA_id |
+------------+-----------+---------+----------+
| Q42338     |         2 |      88 | B12D     |
| O22414     |         3 |      89 | B12D     |
| Q9XHD5     |         3 |      89 | B12D     |
| Q940E1     |        29 |     116 | B12D     |
| Q9LJ47     |         1 |      87 | B12D     |
+------------+-----------+---------+----------+
5 rows in set (0.00 sec)

Annotation: Information displayed on family page

  • Pfam annotation
select * 
from pfamA , pfamA_web 
where pfamA_id = "CBS" and pfamA.auto_pfamA = pfamA_web.auto_pfamA;
  • Interpro annotation
SELECT interpro_id, abstract 
FROM   interpro AS i,
       pfamA AS p
WHERE  p.auto_pfamA = i.auto_pfamA AND
       pfamA_id = "CBS";
  • Gene Ontology (GO) annotation
SELECT go_id, term, category
FROM   gene_ontology AS go,
       pfamA AS p
WHERE  go.auto_pfamA = p.auto_pfamA AND
       pfamA_acc = "PF00067";
  • Literature references
select pfamA_literature_references.comment, order_added, medline, title, literature_references.author , journal 
from pfamA, pfamA_literature_references, literature_references 
where pfamA_id = "CBS" and pfamA.auto_pfamA = pfamA_literature_references.auto_pfamA
and pfamA_literature_references.auto_lit = literature_references.auto_lit
  • Database References
select db_id, pfamA_database_links.comment , db_link, other_params 
from pfamA, pfamA_database_links 
where pfamA_id = "CBS" and pfamA.auto_pfamA = pfamA_database_links.auto_pfamA

Structures and domain interactions (iPfam)

  • Tables containing PDB information

There are two tables containing information primary about PDB structures: pdb and msd_data. The pdb table contains a list of PDB identifiers, the header and the title records from that PDB file.

select pdb_id, header, title from pdb where pdb_id="2abl";
+--------+--------------+-----------------------------------------------------------+
| pdb_id | header       | title                                                     |
+--------+--------------+-----------------------------------------------------------+
| 2abl   | Transferase  | Sh3-sh2 domain fragment of human bcr-abl tyrosine kinase  |
+--------+--------------+-----------------------------------------------------------+
1 row in set (0.00 sec)

For the PDB information to be useful to Pfam we need to map between PDB residues and UniProt sequence residuess. This is not a trivial task! This mapping information is provided by the MSD database. See here for more details. The msd_data table contains this residue by residue mapping.

The following statement gets the first 10 residue mappings for the structure 2ABL.

mysql> select pdb_id, pdb_res, pdb_seq_number, pfamseq_acc, pfamseq_res, pfamseq_seq_number 
       from msd_data, pdb, pfamseq 
       where pdb.auto_pdb=msd_data.auto_pdb and pfamseq.auto_pfamseq=msd_data.auto_pfamseq and pdb_id="2abl" 
       limit 10;
+--------+---------+----------------+-------------+-------------+--------------------+
| pdb_id | pdb_res | pdb_seq_number | pfamseq_acc | pfamseq_res | pfamseq_seq_number |
+--------+---------+----------------+-------------+-------------+--------------------+
| 2abl   | MET     |             75 | P00519      | A           |                 56 |
| 2abl   | GLY     |             76 | P00519      | G           |                 57 |
| 2abl   | PRO     |             77 | P00519      | P           |                 58 |
| 2abl   | SER     |             78 | P00519      | S           |                 59 |
| 2abl   | GLU     |             79 | P00519      | E           |                 60 |
| 2abl   | ASN     |             80 | P00519      | N           |                 61 |
| 2abl   | ASP     |             81 | P00519      | D           |                 62 |
| 2abl   | PRO     |             82 | P00519      | P           |                 63 |
| 2abl   | ASN     |             83 | P00519      | N           |                 64 |
| 2abl   | LEU     |             84 | P00519      | L           |                 65 |
+--------+---------+----------------+-------------+-------------+--------------------+
10 rows in set (0.00 sec)

Using a similar query to the previous one, we generate a mapping for each Pfam domain with a known structure. This information is stored in the pdbmap table.

select pdb_id, chain, pdb_start_res, pdb_end_res 
from pdb, pdbmap, pfamA 
where pfamA_id = 'CBS' and pfamA.auto_pfamA = pdbmap.auto_pfam 
and pfam_region = '1' and pdbmap.auto_pdb = pdb.auto_pdb
  • Tables Containing Domain Interaction Information - i.e. iPfam

iPfam is a database within a database. iPfam contains information about domain-domain interactions.

The main table for iPfam is the interaction table. This is a large denormalised table that contains all of the interaction information to the residue-residue level. The interaction table can be joined onto the following tables: pdb, pfamA, pfamseq, int_atom, int_pfamAs

select distinct interaction.auto_pfamA_A, interaction.pfamA_id_A, interaction.auto_pfamA_B, interaction.pfamA_id_B
from interaction, pdb 
where pdb_id="2abl" and interaction.auto_pdb=pdb.auto_pdb;

There are three other tables that are part of the iPfam specific tables. The int_atom contains the in atom numbers that are forming the interaction. The int_bond table contains the actual bond that is formed between the interaction.

mysql> select pdb_id, interaction.pdb_seq_number_A, pfamseq_seq_number_B, int_atom.pdb_atom, int_atom.partner_pdb_atom 
       from interaction, pdb, int_atom
       where pdb_id="2abl" and int_atom.auto_atom_int=interaction.auto_atom_int and interaction.auto_pdb=pdb.auto_pdb
       limit 10;

Finally, the int_pfamAs table gives the listing of the Domain-Domain interactions found in iPfam.

Genomes

  • ncbi_code 1423 is for species: Bacillus subtilis. This information if found in the ncbi_taxonomy table *

Return all the species & basic Pfam information for a Kingdom:

select ncbi_code, species,  num_distinct_regions, num_total_regions, num_proteins ,
	     sequence_coverage , residue_coverage, total_genome_proteins  
from genome_species 
where grouping like '%Bacteria%' order by species;
  • Return all the Pfam-A domains for a species (using ncbi codes)
select genome_seqs.auto_pfamA, pfamA_acc, pfamA_id, description, sum(count) 
from  genome_seqs, pfamA 
where  genome_seqs.ncbi_code = '1423' and genome_seqs.auto_pfamA = pfamA.auto_pfamA 
      group by genome_seqs.auto_pfamA;
  • Return the protein sequences for a species
select pfamseq.pfamseq_id 
from pfamseq, genome_seqs 
where ncbi_code = '1423' and  genome_seqs.auto_pfamseq = pfamseq.auto_pfamseq;
  • Return all the protein sequences for a species and a specific Pfam-A domain
select pfamseq.pfamseq_id 
from pfamseq, genome_seqs, pfamA 
where ncbi_code = '1423' and  genome_seqs.auto_pfamseq = pfamseq.auto_pfamseq 
and genome_seqs.auto_pfamA = pfamA.auto_pfamA and pfamA_acc = 'PF00106'

See also

  • Integr8 — Access to complete genomes and proteomes
  • TrEMBL Database performing an automated protein sequence annotation
  • InterPro Integration of protein domain and protein family databases

References

  • Finn RD, Mistry J, Schuster-Bockler B, Griffiths-Jones S, Hollich V, Lassmann T, Moxon S, Marshall M, Khanna A, Durbin R, Eddy SR, Sonnhammer EL, Bateman A (2006). "Pfam: clans, web tools and services". Nucleic Acids Res, 34:D247-D251; PMID 16381856.
  • Finn RD, Marshall M, Bateman A (2005). "iPfam: visualization of protein-protein interactions in PDB at domain and amino acid resolutions". Bioinformatics, 21:410-412; PMID 15353450.
  • Bateman A, Coin L, Durbin R, Finn RD, Hollich V, Griffiths-Jones S, Khanna A, Marshall M, Moxon S, Sonnhammer EL, Studholme DJ, Yeats C, Eddy SR (2004). "The Pfam protein families database". Nucleic Acids Res, 32(Database issue):D138-D141; PMID 14681378.

External links