Template-guided assembly
3 methods · 2010–2022
Template-guided assembly: De novo peptide reads assembled into full-length proteins against a related reference, which supplies the order of the pieces and fills the gaps the reads cannot cover.
De novo peptide reads assembled into full-length proteins against a related reference, which supplies the order of the pieces and fills the gaps the reads cannot cover.
The earliest of its 3 methods is Template proteogenomics (2010); 2 more have followed.
| Methods | 3 |
| Papers describing them | 3 |
| Authors | 20 |
| Active | 2010-06-01 to 2022-10-01 |
| Deep learning | 1 of 3 |
| Kinds | adjacent, algorithm, post-processor |
| Acquisition | DDA (2) |
Methods (3)
Oldest first, by the paper that describes each one.
- Template proteogenomics (2010): Reconstructs full target protein sequences, including mutations, from MS/MS spectra by using a homologous protein as a template where the database is imperfect.
- Supernovo (2017): Automated antibody de novo sequencing: finds the closest-matching germline V-J-C sequences by database search, then converges on the true heavy and light chains by iterative wildcard substitution against the MS/MS spectra. Commercial, shipped in Protein Metrics’ Byos platform.
- Deep learning protein scaffold filling (2022): CNN and LSTM models trained on BLAST homologs predict the missing amino acids in a de novo protein scaffold, tested on alemtuzumab.
Papers describing them (3)
- Template Proteogenomics: Sequencing Whole Proteins Using an Imperfect Database (2010, Molecular & Cellular Proteomics, peer-reviewed)
- Automated Antibody De Novo Sequencing and Its Utility in Biopharmaceutical Discovery (2017, Journal of the American Society for Mass Spectrometry, peer-reviewed)
- Deep Learning Approaches for the Protein Scaffold Filling Problem (2022, 2022 IEEE 34th International Conference on Tools with Artificial Intelligence (ICTAI), peer-reviewed)