Learning-to-rank
2 methods · 2019–2020
Learning-to-rank: A ranking model over candidate peptides. The candidates come from a sequencer upstream, and the contribution is putting the right one on top, usually with a predicted spectrum to compare against.
A ranking model over candidate peptides. The candidates come from a sequencer upstream, and the contribution is putting the right one on top, usually with a predicted spectrum to compare against.
The earliest of its 2 methods is pNovo 3 (2019).
| Methods | 2 |
| Papers describing them | 3 |
| Authors | 9 |
| Active | 2019-07-24 to 2020-01-01 |
| Kinds | algorithm, post-processor |
| Acquisition | DDA (2) |
Methods (2)
Oldest first, by the paper that describes each one.
- pNovo 3 (2019): Learning-to-rank + pDeep
- GP-based scoring function for de novo PSM re-ranking (2019): Genetic-programming-derived scoring function for re-ranking peptide-spectrum matches emitted by de novo sequencing tools. Improves top-k accuracy over the default scorer.
Papers describing them (3)
- pNovo 3: precise de novo peptide sequencing using a learning-to-rank framework (2019, Bioinformatics, peer-reviewed)
- Improving the Results of De novo Peptide Identification via Tandem Mass Spectrometry Using a Genetic Programming-based Scoring Function for Re-ranking Peptide-Spectrum Matches (2019, arXiv, preprint)
- Evolutionary Algorithms for Improving De Novo Peptide Sequencing (2020, thesis)