GlycopeptideID - intact glycopeptide identification with CID MS2 spectra

NOTE! This is new release candidate. You may use this, but there may still be some changes and service breaks. The current official release is found here

GlycopeptideID is an open access web tool developed to ease the analysis of the glycopeptide CID MS2 spectra. The tool is developed and hosted by Applied Numerics in collaboration with the University of Helsinki.

Key features:
  • automated identification of both peptide sequence and glycan structure or composition
  • user adjustable peptide databases, from a single protein to the whole UniProt DB with desired taxonomy
  • user adjustable glycan databases or de novo search
  • possibility to include peptide and/or glycan modifications
  • probability based scoring
  • false discovery rate (FDR) estimation for peptide sequences
  • capable to process 1000s of MS2 spectra per LC-MS/MS2 run
  • example MS2 data and databases included
  • data storage in the server
  • data download with Excel compatible files
More details are found at the Help page.

Usage

The service is open for guests and registered users. For guests the imported data and results are stored for one week and the total disk quota is limited to 3 GB. For registered users the limits can be negotiated. Also, if large scale analysis is required, set up of a dedicated server or a cluster is possible. To register please contact hannu.peltoniemi(at)appliednumerics.fi.

For smooth usage the service is run in a large (virtual) server. To reduce the costs the server is stopped if the service is not used. If the server is stopped, the restart page is located at the same web address as the running service.

Feel free to use the service. However, the service is currently at beta level and requires more testing with practical use cases. Please, report all bugs. Also, all suggestions for improvements are welcome.
The service is tested with Firefox. Other browsers may have some problems, especially IE. For details of known issues see Notes.

Reference

The official publication of this service is yet to come. However, the computational methods follow closely the methods published at Joenväärä, S. et al., N-Glycoproteomics – an automated workflow approach., Glycobiology 2008 18(4):339-349 and at Peltoniemi, H. et al., Novel data analysis tool for semiquantitative LC-MS-MS2 profiling of N-glycans. Glycoconj. J. 2013, 30(2):159-170.

Contact

For issues about this web service contact Hannu Peltoniemi (hannu.peltoniemi(at)appliednumerics.fi).

Disclaimer

The service is released with a hope that it will be beneficial for glycopeptide analysis. However, Applied Numerics do not take any responsibility of the imported user data or the results given by the service.

Acknowledgements

Several open source or open data projects have been used with this service, including:

Release Notes

2014-11-18, version 0.93 beta

Modifications:
  • Glycan identification without peptide is included.
    If glycan option 'Search glycans also without matching peptide' is selected then also hypotetical peptides are included. A hypotetical peptide is defined by total mass only, and it is required that the data contains a peptide+HexNAc peak. Hypotetical peptide is not included if there is a peptide sequence which matches to that peptide+HexNAc peak.
  • Glycan scoring modified.
    New scoring includes intensity by calculating: What is the probability that by random one would select as many high intensity peaks (intensity above 5% of base peak) as is observed with the given matched glycopetide.
    The old scoring was: What is the probability that by randomly selecting peaks one would cover as much intensity as the matched glycopeptide.
    The difference between the two scoring types should be minor, expect that the new one is much faster.
  • New option: Min intensity limit.
    Removes the MS2 peaks which have intensity (relative to base peak) less than the given value.
  • Intermediate peptide output removed. The aim is to save disk space and increase speed.
  • Output spectrum text files are changed form xml to text (tab delimited). The aim is to save disk space.
  • Improved mgf format support.

2014-02-28, version 0.91 beta

Fixes and modifications:
  • Target-decoy validation changed to filtered FDR.
  • Optional tagging and filtering with typical NeuAc tag peaks.
  • Help updated.

2014-01-30, version 0.9 beta

Major update:
  • Possibility to identify glycan structures.
  • New page Glycan DB for generating glycan databases.

2013-09-22, version 0.81 beta

Minor fixes and modifications:
  • Guest disk quota limit set to 3 GB.
  • Used user disk quota added to Data page.

2013-05-15, version 0.8 beta

Web pages:
  • Major update to all web pages.
  • Results matrix style is modified. New style affects only new results, old results look more or less the same as before.
Known issues:
  • Option validation doesn't work properly with Internet Explorer (IE). Some option boxes are marked to red even when they are OK. The bug doesn't affect the calculation and the tool can be run normally. At the moment the best fix is to avoid IE.
  • Possible inconvenience with service restart:
    The restart script sends a start command and waits a while. It do not actually detect that the server is running. In some cases the wait time is not long enough and the restart page is returned when user clicks search, results, peptide db or data page. If that happens wait a second and try again. No new restart is required.