Proteoform search from protein database with top-down mass spectra.

Area of Science:

• Proteomics
• Bioinformatics
• Computational Biology

Background:

• Accurate identification of proteoforms is crucial for understanding protein function and biological processes.
• Existing computational methods for proteoform identification can be slow and may lack sufficient accuracy.

Purpose of the Study:

• To develop a novel, efficient, and accurate search algorithm for proteoform identification from mass spectrometry data.
• To improve upon the speed and accuracy of current proteoform identification tools.

Main Methods:

• Proposed a search algorithm that computes largest-size error-correction alignments between protein and spectrum mass graphs.
• Developed a combined method utilizing filtering and search algorithms for candidate identification and result reporting.
• Created a pipeline for generating simulated top-down spectra from protein sequences with modifications.

Main Results:

• The exact searching method demonstrated a 3.9 to 9.0 times speed improvement over popular methods like TopMG and TopPIC.
• The combined method accelerated the running time of sTopMG without compromising search accuracy.
• Experiments on simulated datasets yielded 95% accuracy, surpassing existing methods.
• Real annotated datasets achieved ≥97.1% accuracy when using the FLASHDeconv deconvolution method.

Conclusions:

• The proposed search algorithm offers a significant advancement in the speed and accuracy of proteoform identification.
• This method provides a robust computational tool for analyzing complex proteomic datasets.
• The developed pipeline and algorithms are valuable for both simulated and real-world proteomic research.