Spectral Similarity Prediction Evaluation

---
name: spectral-similarity-prediction-evaluation
description: Use when use when the workflow requires spectral-similarity-prediction-evaluation.
license: CC-BY-4.0
metadata:
  edam_topics: []
  tools:
  - MS2DeepScore
  - RDKit
  - Python
  - scikit-learn
  license_tier: open
  provenance_tier: literature
derived_from:
- doi: 10.1186/s13321-021-00558-4
  title: MS2DeepScore
evidence_spans:
- Our MS2DeepScore Python library offers two types of data generators
- To estimate the uncertainty of a prediction we used Monte-Carlo Dropout ensembles
- Unless noted otherwise, we used Tanimoto scores on RDKit [23] Daylight fingerprints
  (2048 bits) to compute structural similarities.
- Unless noted otherwise, we used Tanimoto scores on RDKit [23] Daylight fingerprints
  (2048 bits) to compute structural similarities
- Our MS2DeepScore Python library offers two types of data generators, one which iterates
  over all unique InChIKeys (DataGeneratorAllInchikeys) and one which iterates over
  all spectra and was used for
- Using the t-SNE [28] implementation from scikit-learn [29] we computed two-dimensional
  coordinates
claims: []
provenance:
  collection: https://w3id.org/holobiomicslab/asb-skill/collection/metabolomics/v2
  assembled_by: scripts/collect_metabolomics_collection.py
  sources:
  - build: coll_ms2deepscore_cq
    doi: 10.1186/s13321-021-00558-4
    title: MS2DeepScore
  dedup_kept_from: coll_ms2deepscore_cq
schema_version: 0.2.0
attribution:
  generator: AgenticScienceBuilder
  original_doi: 10.1186/s13321-021-00558-4
  all_source_dois:
  - 10.1186/s13321-021-00558-4
  zenodo_doi: 10.5281/zenodo.20794027
  curators: []
  promoter: Louis-Félix Nothias
  sponsor: CNRS & Université Côte d'Azur
---

# spectral-similarity-prediction-evaluation

## When to use

Use when the workflow requires spectral-similarity-prediction-evaluation.