Welcome to the DPInteract

• Usage

  Welcome to the DPInteract! We have not gone public with this data yet, and so please do not pass the URL around. If you know of anyone who might find this dataset useful, please let us know. This is mostly so we know who to warn of changes. Please let us know if you plan to publish a paper using data from this dataset.

  Robison, K., Church G.M. (1994). The DPInteract. Unpublished

  This database is copyright 1994, the President and Fellows of Harvard College.

• Purpose

  This dataset is being collected with several purposes in mind
  1. Cataloging demonstrated sites and non-sites for E.coli DNA-binding proteins
  2. Aiding the annotation of such sites in other E.coli databases and sequence entries
  3. Interpreting the results of whole-genome in vivo methylation protection experiments (Nature 360: 606-610; J Bacteriol 176: 3438-3441)
  4. Developing better computational tools for recognizing DNA binding proteins in sequence data

• Access

  All access is currently through the World Wide Web (WWW), using a browser such as Mosaic, NetScape, or Lynx. We currently offer 3 ways of finding data in the database

  Outline

  Many entries are categorized into families, which are further organized into structural classes. The top page of the dataset WWW server displays these classes and the gene families contained in them. Picking a family will produce an entry listing all genes in that family

  Full-text search

  You may search the text annotations by keywords.

  Citation search

  You may search for citations within the dataset

  BLASTN search

  A form is provided to search a DNA sequence against the database using BLASTN. This is intended to be used to look-up recorded sites in a DNA sequence; it is not expected to be useful for identifying new sites.

  Matrix search

  A form is provided to search a DNA sequence against matrices for specific sites. These matrices have been prepared from sites in the database, and calibrated on the E. coli genome.

  Sequence search

  You can search for a particular short sequence. This is intended to be used to look for restriction/methylation sites. Please remember the "edge effect" -- the search is working on the data in the database, and not the whole sequence, and so a restriction site which is only partially represented in the database will be missed (e.g. If the original sequence was TTTGAtc, and only the letters in caps are in our database, then this search will miss the Dam site).

  Scope

  In order to keep this project manageable, we are targeting our efforts towards the E.coli chromosome.
  1. Known or suspected (HTH-bearing or family membership) sequence-specific DNA-binding proteins from the Escherichia coli K12 chromosome
  2. Known or suspected sequence-specific DNA-binding proteins from Salmonella typhimurium LT2
  3. Well-studied phage sequence-specific DNA-binding proteins (e.g. cro)
  4. Well-studied non-E.coli proteins from large gene families, which may yield insight into E.coli proteins of the same family (e.g. luxR from Vibrio, nodD from Rhizobia [lysR family])
  Some data on other proteins may have been collected as a by-product of literature-scanning efforts. However, only for E.coli proteins will an attempt at completeness be made.

• Cross-Links

  Most of the important contextual information for these proteins is not directly in the database. Instead, we have provided hyperlinks to a number of other databases.
  • Protein Sequence. Where possible, we cross-reference SwissProt. Otherwise, we pick PIR, or at worst, GenPept.
  • Protein Sequence Patterns. Where available, the gene family records contains links to Prosite and BLOCKS.
  • Citations. Medline citations are retrived from the Entrez Molecular Sequences subset when present there, and otherwise from the Genetics Subset. SeqAnalRef is used for citations not found in Medline.
  • Genetic Information -- the Coli Genetic Stock Center (CGSC)'s database.

• Status

  We believe we have captured all of the footprinted E.coli DNA binding sites published before May 1996. We would be very interested to find any sites missing from the database. At this time, rpoD (sigma70) has been largely excluded from this phase, since other workers have compiled these sites.

  Efforts are currently focused on:

  • Maintaining completeness via scanning the current literature
  • Continued identification of cross-references, particularly Medline xrefs for recent papers and CGSC gene xrefs.
  • Creating alignments, matrices, and logos of all binding sites
  • Comparing predicted binding sites vs observed, using several prediction methods.
  • Finding citations for comparisons of binding sites between E.coli and other bacteria
  • Finding citations for insertional or deletional knock-outs of DNA-binding proteins
  • Finding citations for altered-specificity mutants of DNA-binding proteins

DPInteract Home