* SPD DataSet Release On Aug.11 2004 * 

spd.all.gz: 
  description: All SPD protein sequences in fasta format.
  format:      Header line (protein_id, name if exists, description, species, data source, length), sequence line.
spd.nr90.gz       
  description: A non-redundant dataset of "spd.all.gz". Via pairwise-blastp, sequences with overall-identity >= 90% and overall-coverage (the ratio between the tiled HSPs and the short one between query and subject) >= 90% are excluded.
  format:      Identical with "spd.all.gz".
spd.nr90.species.gz             
  description: In comparison with the above "spd.nr90.gz", redundancy is removed only in the same species, i.e., similar sequences among different organisms are remained.
  format:      Identical with "spd.all.gz".

spd.mrna.gz
  description: Corresponding mRNA sequences for all entries if exist in fasta format.
  format:      Header line (mRNA accession number, protein_id, length), sequence.

spd.name.gz	
  description: A list of spd protein names if exists (tab-delimited).
  format:      Protein_id, official name, rank, species (ordered by species and rank).
spd.name.nr90.gz
  description: The version of "spd.name.gz" corresponding to "spd.nr90.gz".
  format:      Identical with "spd.name.gz".
spd.name.nr90.species.gz
  description: The version of "spd.name.gz" corresponding to "spd.nr90.species.gz".
  format:      Identical with "spd.name.gz".

spd.ref.gz
  description: Reference(paper) number of each proteins in SPD (tab-delimited).
  format:      Protein_id, number.

spd.classes.domain.gz
  description: Representative domains used to classify SPD core proteins (tab-delimited). 
  format:      Domain accession number, domain name, functional category (ordered by category, accession number).
spd.classes.pro.gz
  description: Classification of SPD core proteins with known SwissProt secreted proteins excluded via domain analysis (tab-delimited).
  format:      Protein_id, functional category (ordered by category, protein_id)

spd_core_ref.mcl.gz
  description: Cluster of SPD core proteins as well as nine reference datasets via tribe-mcl
  format:      cluster_id, protein_id, protein source, member number of cluster
  notes:       Some proteins might have multiple data sources. For example, both GO derived secreted proteins and SwissProt vertebrate secreted proteins contains "MM20_BOVIN", "WN3A_XENLA", etc.


For further information, please refer to http://spd.cbi.pku.edu.cn/help/spd_help.php

Center of Bioinformatics, Peking University