Estimating probabilities of peptide database identifications to LC-FTICR-MS observations
2006

Improving Peptide Identification in Proteomics

Sample size: 32 publication Evidence: moderate

Author Information

Author(s): Kevin K. Anderson, Matthew E. Monroe, Don S. Daly

Primary Institution: Pacific Northwest National Laboratory

Hypothesis

Can using a priori probabilities improve the identification of target peptides in proteomics?

Conclusion

Using a priori probabilities that contrast peptides from expected and unexpected proteins improves the identification of target peptides.

Supporting Evidence

  • A total of 69,220 features were analyzed across 32 LC-FTICR-MS analyses.
  • High confidence identifications were achieved with a conditional probability greater than 0.95.
  • Using contrasting prior probabilities improved identification rates compared to equally likely prior probabilities.

Takeaway

This study shows that using specific probabilities helps scientists find the right pieces of proteins in a sample, making it easier to understand what’s there.

Methodology

The study used LC-FTICR-MS to analyze a tryptically digested bovine serum albumin sample and matched the results to a peptide database using statistical classification methods.

Potential Biases

The presence of false positives from unexpected sources may lead to misidentifications.

Limitations

The completeness of the PMT tag database is an issue, as 82.3% of the features were unmatched.

Statistical Information

P-Value

0.95

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1477-5956-4-1

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication