Restricted Data Portal Added to GPMDB (2008/11/19)

Since the inception of the GPM system, GPM search servers have offered investigators the option of storing their results in the the public data repository GPMDB. Starting today, an additional choice has been made available for investigators that would like their results recorded in GPMDB, but would like to restrict access to those results. This restricted access option is now available on all search submission pages. Access to this restricted data can be obtained through the The GPMDB Portal for Restricted Access Data Sets which can be used by the investigators for the purposes of manuscript review or simply to maintain the privacy of their results. A short description of the portal and its use is available on the GPM wiki.

HUPO Plasma Proteome Project 2 Data Release (2008/10/28)

The most recent data from the Human Proteome Organization's Plasma Proteome Project released into Tranche has been added to GPMDB, as part of the ProteomExchange system. This release is comprised of 485 data sets, generated by three laboratories. To view these data sets, use the data set keyword HUPO PPP2. To view all data sets identified as being from plasma, use the Brenda Tissue Ontology accession number for blood plasma as the data set keyword BTO:0000131.

New PTM files available (2008/10/21)

As an ongoing part of the GPM Tornado project, new post-tranlational modification annotation files are now available from the GPM ftp site. These annotation files are based on Release 56.3 of the SwissProt Protein Knowledgebase and ENSEMBL version 50 protein sequences for the following species:

1. Homo sapiens (human_mod.xml);

2. Mus musculus (mouse_mod.xml);

3. Gallus gallus (chicken_mod.xml);

4. Rattus norvegicus (rat_mod.xml); and

5. Saccharomyces cerevisiae (yeast_mod.xml).

New species included in MRM worksheet (2008/10/14)

The GPM's Multiple ion Reaction Monitoring Worksheet has been extended to Mus musculus and Saccharomyces cerevisiae protein and peptide sequences. The original release could only be used with H. sapiens sequences. Links to the appropriate worksheet pages can be found in the list of links at the top of protein display pages for all ENSEMBL mouse and SGD yeast accession numbers. These links are positioned as shown in this example (mouse prosaposin):

Version 3 of the NCTA protein lists (2008/10/14)

The 3rd release of the Normal Clinical Tissue Alliance's tissue specific lists of observeable proteins is now available. This release includes a new tissue, liver, in addition to the original 13 tissue types. The underlying data has been re-curated and the protein accessions aligned with the most recent release of ENSEMBL (v. 50).

2008.10.01 release of the X! Hunter ASLs (2008/10/10)

The 2008.10.01 version of the X! Hunter Annotated Spectrum Libraries has been released. This new version incorperated some significant new data sets, such as Steve Carr's tissue specific mouse mitochondrial proteomes; the HUPO Liver Proteome Project and HUPO Plasma Proteome Project 2. See the project page for details.

Updated versions of distributed Parallel Tandem for PVM and MPI are now available. (2008/09/08)

Both have been tested for 32 and 64 bit kernels on AMD Opteron and Power PC (ftp). This update places together in one distribution the most recent programs that have been tested for both 32 and 64 bit compatibility. There are no other new features in this update.

Changes to the "peptide" display (2008/09/03)

The display controls for the "peptide" page in the GPM have been changed to give the user faster access to information about peptide modifications for a particular data set. The new control is shown here:

The hyperlinks on the original page that allowed the display to change from displaying all peptides to either the best, mutant, non-tryptic or modified peptides have been consolidated into a single drop-down menu. A new control ("Find") that allows the display to only show peptides that contain a particular peptide sequence is located beside this drop-down menu. A summary table of all of the amino acid residue modifications that have been assigned in the data set is now shown just below the new controls. By clicking on the links in that table, either all of the peptides with a particular modification mass or the peptides with a particular mass/residue type combination can be displayed. The numbers in square brackets are the total numbers of residues modified.

New X! Hunter spectral libraries (2008/08/28)

A new set of mass spectrum libraries for use in protein identifications has been made available. Check the X! Hunter library page for details. The library curation process has been improved to improve the accuracy of the spectra stored in the libraries. Several new species have been added (horse and fugu), as well as significant expansion of the zebrafish and arabidopsis libraries because of large new data sets added to GPMDB.

X!!Tandem now available (2008/08/08)

Announcing the availability of X!!Tandem, a version of X!Tandem parallelized for running on clusters or networks of workstations.

X!!Tandem is a parallel, high performance version of X!Tandem that has been parallelized via MPI to run efficiently on large numbers of CPUs. In X!!Tandem the search is parallelized by splitting the input spectra into as many subsets as there are processors, and processing each subset independently. Both compute-intensive stages of the processing (initial and refinement) are parallelized, and overall speedups in excess of 20-fold have been observed on real datasets. With the exception of minor details related to MPI's program launching, X!!Tandem is run in exactly the same manner as X!Tandem, using the same input and configuration files, and produces exactly the same output.

X!!Tandem is a project of Yale's Keck Biotechnology Resource Laboratory, and was developed by Robert Bjornson (robert.bjornson@yale.edu). The package can be downloaded from here.

ENSEMBL 50 update (2008/08/01)

The protein sequences in the public GPM have been updated to the recently released ENSEMBL 50 sequences. These updates also include new SNAP data for human, mouse and rat sequences.

Indexing of human, mouse and rat SNAP data (2008/07/21)

The data collected in GPMDB that demonstrates the presence of proteins that contain Single Nuclotide-induced Amino acid Polymorphisms (SNAPs) caused by known non-synomymous SNPs has been indexed for easier use. In order to find the peptides that have been observed with an amino acid polymorphism corresponding to a particular SNP, simply enter the SNP accession number into the accession number form (e.g., rs2692696, rs1800215).

Accession number translation feature (2008/06/23)

A new feature that allows alternative translations for human and mouse ENSEMBL protein accession numbers is now available. Any model originally obtained using ENSP or ENSMUSP accessions can now be displayed using the following alternative annotations, without having to search the data again. The following on-the-fly translations are now available (click on the link to see the same data annotated with alternative accessions):

1. ensembl - default;

2. ipi - International Protein Index;

3. symbol - HGNC or MGI gene symbols; or

4. ncbi - NCBI gene accession numbers;

5. swiss - Swiss-Prot/Uniprot accession numbers.

This translation can be performed by selecting the appropriate value from the "model" page drop-down list. These translations are approximate, as there may be no completely accurate translation between various sequence sets, but the translation is as accurate as we can make it.

50,000,000th Peptide Id Recorded (2008/05/16)

Earlier this week, GPMDB passed the 50 million mark for peptide identifications. We would like to thank all of the data contributors who have made this project a success.

Experimental MRM prediction interface (2008/05/16)

A new interface that attempts to predict multiple-ion reaction monitoring transitions for peptide sequences has been made available, for selected human proteins and peptides (details). The prediction algorithm is based on the Annotated Spectrum Library data set, used by X! Hunter for protein identifications. It attempts to find unique transitions for a given peptide sequence and parent ion m/z, given parent and fragment ion mass tolerances. The format of the output page will change as users give us suggestions to enhance its utility. The prediction algorithm is available through the new | mrm | links that occur on the link toolbars on protein and peptide display pages.

Additions to GPMDB interface (2008/05/09)

Several new ways of accessing information in the GPMDB have been added.

Normal Clinical Tissue Alliance site opens (2008/03/24)

The Normal Clinical Tissue Alliance was formed to provide information about the proteins that can be observed in clinically derived tissues. Lists of proteins for each of the tissue types can be obtained and manipulated. The tissues are "normal" (non-diseased) and the lists represent composites of proteins obtained from multiple studies. The protein lists are dependent on the availability of proteomics data: the lists are compiled directly from MS/MS identification data, which is also made available through the site.

Change to protein coverage display (2008/03/23)

The protein sequence coverage displays have conventionally showed the regions of a protein that have been observed in red. The displays have been updated to show the regions of a protein that are predicted to be difficult to observe (using MS/MS-based proteomics) in green.

Species added to GPM servers (2008/03/12)

Twenty-two new eukaryote species have been added to the GPM Tornado cloud server system. These species include 14 new fungi, 2 insects and 6 protists. A list of these new species is as follows:

1. Aspergillus nidulans FGSC A4
2. Candida albicans
3. Candida glabrata CBS138
4. Cryptococcus neoformans var JEC21
5. Debaryomyces hansenii CBS767
6. Encephalitozoon cuniculi
7. Eremothecium gossypii
8. Gibberella zeae
9. Kluyveromyces lactis NRRL Y-1140
10. Magnaporthe grisea 70-15
11. Pichia stipitis
12. Ustilago maydis
13. Yarrowia lipolytica
14. Yarrowia lipolytica CLIB122
15. Nasonia vitripennis
16. Tribolium castaneum
17. Cryptosporidium parvum
18. Dictyostelium discoideum
19. Leishmania infantum
20. Plasmodium falciparum
21. Theileria parva
22. Trypanosoma brucei

These new proteomes were all obtained from NCBI.

GPM changes to a peer-to-peer grid computing system (2008/02/25)

Starting on Feb. 23, 2008, a peer-to-peer grid computing system (called Tornado) is being rolled out for the GPM search servers. When a search is submitted to any of the genome, human, mouse or rat search pages, the system will determine which server on the grid is the fastest and least busy and send your search to that computer. This system has the effect of increasing the effective capacity of the existing system by approximately 10-fold.

Note: In order to use this system, the search form must say "GPM Tornado" at the top of the page. If it does not say Tornado, please press the reload button on your browser to get the most recent version of the search form.

KEGG pathways analysis added (2008/01/24)

A new display using the Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathways database is now available for all data sets that have used the ENSEMBL proteomes for human, mouse, rat or yeast. The new display categorizes the identified proteins by metabolic pathway, with about 190 pathways available for human proteins. The pathways are linked through to their KEGG cartoon representations.

Annotated modifications added (2008/01/20)

Starting with the 2008.02.01 release of X! Tandem, the search engine will now have the capacity to set the potential modifications being tested on a protein by protein basis. This new feature can be activated in either the first round of searching or the subsequence refinement rounds by setting the "Use sequence annotations" control to be "yes".

The way that this new feature works is quite simple. A file is constructed that contains a list of sequence accession numbers and potential sequence modification specifications, e.g. several lines from the human modification file look like the following:

<protein label="ENSP00000166244" pmods="79.966331@Y" />
<protein label="ENSP00000350614" pmods="79.966331@S" />
<protein label="ENSP00000372956" pmods="79.966331@S,79.966331@T" />
<protein label="ENSP00000372947" pmods="79.966331@S,79.966331@T" />
<protein label="ENSP00000363773" pmods="15.994915@P" />

The first line indicates that the protein sequence ENSP00000166244 is known to be tyrosine phosphorylated. If the "Use sequence annotations" feature is turned on, then that sequence will be tested for tyrosine phosphorylation, in addition to the other potential modifications you have specified for your search to be applied to all protein sequences. Similarly ENSP00000350614 will be checked for serine phosphorylation, ENSP00000372956 and ENSP00000372956 will be checked for phosphorylation at serine and threonine residues, while ENSP00000363773 will be tested for the possible presence of hydroxyproline residues.

Files of this type have been constructed for human, mouse, rat, chicken and brewer's yeast proteomes, using publically available annotation sources such as Uniprot and GPMDB. This capability is now available on all of the public GPM search servers. The annotation files are available here.

GPM system updates (2008/01/06)

Over the Holidays, several system changes have been made to the GPM system and interface pages. While most of these changes are minor, they do change some default behaviors.

1. The "peptide clustering" control on the protein page has been changed to be consistent with the similar control on the "peptide" page. For both pages, the default setting is now "all", meaning that all identified spectra will be represented on the page. This change was made at the suggestion of a number of users, because of some confusion caused by having only the "best" identification for each peptide sequence shown as the default.
2. More information about the distribution of proteins amongst Gene Ontology categories has been added to the "GO" display, including p-values for enrichment or depletion each category.
3. Shading and other small formatting and navigations changes have been made to make the interface as consistent as possible.