Research map / Probabilistic and scientific modeling / Variational and generative sampling
Learning Concave Conditional Likelihood Models for Improved Analysis of Tandem Mass Spectra
John T Halloran, David M Rocke
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Train Didea with CVE emissions and learn multiple shift-specific parameters | Didea peptide scoring · Presented datasets (average) | spectra identified at 1% FDR: 16% more spectra than DRIP | Improved |
| Train Didea with CVE emissions and learn multiple shift-specific parameters | Didea peptide scoring · Presented datasets (average) | spectra identified at 1% FDR: 17.4% more spectra than MS-GF+ | Improved |
| Train Didea with CVE emissions and learn multiple shift-specific parameters | Didea peptide scoring · Presented datasets (average) | spectra identified at 1% FDR: 9.8% more spectra than Didea-0 | Improved |
| Replace DRIP gradient features with Didea conditional log-likelihood gradients | Percolator recalibration · Presented datasets (average) | spectra identified at 1% FDR: 12.3% more spectra than DRIP | Improved |
| Replace DRIP gradient features with Didea conditional log-likelihood gradients | Percolator recalibration · Presented datasets (average) | spectra identified at 1% FDR: 13.4% more spectra than MS-GF+ | Improved |
| Apply the optimized exact sum-product inference schedule | Didea database search · Worm-1 | runtime: 64.2% runtime improvement | Improved |
| Use the optimized inference schedule instead of the Didea-0 implementation | Didea database search · Worm-1 (1,000 randomly sampled spectra) | runtime (seconds per spectrum): 19.1175 -> 6.8535 | Improved |
| Hash virtual-evidence bin values by b-/y-ion value and shift (ablation) | Didea inference · Runtime optimization | runtime | No clear effect |
Closest papers by meaning
- PepSpecBench: A Unified Evaluation Benchmark for Peptide Tandem Mass Spectrometry Prediction (NeurIPS 2026 · similarity 0.88)
- GLACIER: Rethinking Mass Spectrum Prediction as an Object Detection Problem (NeurIPS 2026 · similarity 0.87)
- Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (ICML 2025 · similarity 0.87)
- AdaNovo: Towards Robust \emph{De Novo} Peptide Sequencing in Proteomics against Data Biases (NeurIPS 2024 · similarity 0.86)
- ContraNovo: A Contrastive Learning Approach to Enhance De Novo Peptide Sequencing (AAAI 2024 · similarity 0.85)
- Kernel-Based Evaluation of Conditional Biological Sequence Models (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
