π-HelixNovo2
algorithm · Transformer (AR)
π-HelixNovo2: algorithm · Transformer (AR). Successor to π-HelixNovo with an emphasis on availability: an online inference service alongside the model architecture refinement. Same Tsinghua / Pengcheng Lab / NCPSB…
Successor to π-HelixNovo with an emphasis on availability: an online inference service alongside the model architecture refinement. Same Tsinghua / Pengcheng Lab / NCPSB collaboration as the original.
| Kind | algorithm |
| Deep learning | yes |
| Acquisition | DDA |
| Family | Transformer (AR) |
Code
Not tracked on the Code activity chart: those figures come from the GitHub API, and this link is not a GitHub repository.
Data
What the papers describing this method trained it on and reported its results on, as each paper states it. A version is named only where the paper names one.
Trained on
- BALF proteomics - In-depth proteomic analysis of human bronchoalveolar lavage fluid towards the biomarker discovery for (as deposited)
- BoxCar acquisition method enables single shot proteomics at a depth of 10,000 proteins in 100 minutes (as deposited)
- Casanovo data set and model weights (as deposited)
- Casanovo non-enzymatic fine-tuning set (as deposited)
- Diabetes causes marked inhibition of mitochondrial metabolism in pancreatic β-cells (as deposited)
- HeLa proteome of 12,250 protein-coding genes (as deposited)
- Low-density lipoprotein receptor-related protein 1 (LRP1)-derived peptides protect against aggregation of LDL and choles (as deposited)
- MassIVE-KB (v1)
- MassIVE-KB (v1 30M-PSM split (Casanovo))
- Mining waste streams of food production for bioactive plant polysaccharides that affect the fitness and expressed activi (as deposited)
- NIST peptide tandem mass spectral libraries (version not stated)
- NIST peptide tandem mass spectral libraries (synthetic HCD)
- Nine-species benchmark (original (DeepNovo, 2017))
- ProteomeTools (version not stated)
- Proteomic analysis of six different tissues from the Atlantic bottlenose dolphin (Tursiops truncatus) (as deposited)
Evaluated on
- ABRF PRG 2018 HeLa DDA (PRG 2018 runs)
- Anti-FLAG-M2 antibody sequencing (as deposited)
- Casanovo data set and model weights (as deposited)
- Casanovo non-enzymatic fine-tuning set (as deposited)
- HeLa proteome of 12,250 protein-coding genes (as deposited)
- High-resolution spatially-resolved proteome mapping using automated, sacrificial liquid-mediated sample transfer from la (as deposited)
- Mining waste streams of food production for bioactive plant polysaccharides that affect the fitness and expressed activi (as deposited)
- Monoclonal antibody de novo assembly (as deposited)
- NFYB-1 regulates mitochondrial function and longevity via lysosomal prosaposin (as deposited)
- Nine-species benchmark (original (DeepNovo, 2017))
- Nine-species benchmark (revised (main))
- Predictive signatures of 19 antibiotics-induced Escherichia coli proteomes (as deposited)
- Proteome of the rodent malaria parasite Plasmodium berghei liver stage merosomes (as deposited)
- The Proteome Landscape of the Kingdoms of Life (as deposited)
- The beta subunit of nascent polypeptide associated complex plays a role in flowers and siliques development of Arabidops (as deposited)
Paper describing it
- π-HelixNovo2: Making Accurate Online De Novo Peptide Sequencing Available to All (2026, Genomics, Proteomics & Bioinformatics, peer-reviewed)