Proteoform search from protein database with top-down mass spectra
Bibliographic record
Abstract
Abstract In this paper, we propose a new search algorithm for proteoform identification that computes the largest size error correction alignments between a protein mass graph (PMG) and a spectrum mass graph (SMG). We also design a filtering algorithm. The combined method uses the filtering algorithm to get some candidates and then apply the search algorithm to report the final results. Our exact searching method is 3.9 to 9.0 times faster than the popular methods such as TopMG and TopPIC. Our combined method can further speed up the running time of our method sTopMG by 6.2 times without affecting the search accuracy. Lots of search methods have been developed in the past decade. The search results reported by various methods are known to be significantly different. In literature, there is no top-down mass spectra dataset with known corresponding protein sequences. There is no tool for generating simulated top-down mass spectra with input protein sequences, either. Though there are many published papers to compare and evaluate various search methods, they all use some kind of indirect measures since there is no real or simulated dataset of top-down spectra with known corresponding true protein sequences. Thus, in some sense, the accuracy of existing search methods is somewhat uncertain. Here we develop a pipeline for generating simulated top-down spectra based on input protein sequences with modifications. To our knowledge, this is the first tool to generate simulated top-down mass spectra in a reasonable way indicated by an interesting measure, match gap distribution. Experiments on simulated datasets show that our combined method has of 95% accuracy, while the best existing methods have accuracy far below this. To further evaluation the performance of the existing methods, we generate a set of 55 real top-down spectra from 3 domains of a known antibody. The real dataset shows that our new method has 94.2% accuracy using deconvolution method FLASHDeconv, which is consistent with the accuracy of the simulated data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".