MétaCan
Menu
Back to cohort
Record W2746078860 · doi:10.13034/jsst.v10i1.123

Implementation of virtual workflows in KNIME for medicinal chemistry

2017· article· en· W2746078860 on OpenAlexvenueaboutno aff
Jack Antonio DiTommaso

Bibliographic record

VenueJournal of Student Science and Technology · 2017
Typearticle
Languageen
FieldBiochemistry, Genetics and Molecular Biology
TopicBiomedical Text Mining and Ontologies
Canadian institutionsnot available
Fundersnot available
KeywordsWorkflowComputer scienceData scienceChemistWorld Wide WebChemistryDatabase

Abstract

fetched live from OpenAlex

This project demonstrates how two programs are created in KNIME - an open source data analytic, reporting and integration platform, are used to support research scientists in medicinal chemistry. The first application flags pan-assay interference compounds such as “promiscuous” compounds present in chemical libraries that recurrently behaves as false positive hits in screening campaigns. The second application adapted a previously published workflow, where it automatically scans the recently published scientific literature on a weekly basis, and identifies articles considered relevant to medicinal chemists focused on epigenetic mechanisms, a novel and promising field in drug discovery. These workflows are very important because they allow a user with relatively little training to be able to extract important data that would typically need a trained chemist for. The PAINS workflow performed adequately but data was problematic. This workflow and an online tool, used to compare results, agged different, but overlapping sets of compounds. The PubMed alert workflow performed very well, being able to consistently identify new papers. These workflows have been implemented at the Structural Genomics Consortium, in Toronto. Both Workflows are available at http://sgc.utoronto.ca/ditommaso.zip The implementation of these workflows demonstrate that the process is viable, and paves the way for the implementation of more complex workflows. Ce projet montre comment deux logiciels qui ont été créés en utilisant KNIME - une plate-forme open-source d’intégration et de reportage de data analytique, sont utilisées comme soutient pour les chercheurs dans le domaine de chimie médicale. La première application signale les composés d’interférence pan-essai (PAINS), par exemples des composés ‘libérés’ présents dans les chimiothèques, qui s’agissent souvent comme des fausses réactions positives pendant les campagnes de dépistage. La deuxième application, le système de workflow PubMed alert, a adapté un système de workflow développé auparavant qui parcourt rapidement la littérature scientifique publiée récemment une fois par semaine et identifie des articles qui sont pertinents pour des chimistes médicales qui étudient des mécaniques épigénétiques, un domaine novateur et prometteur dans les découvertes des drogues. Ces systèmes de workflow sont très importants car ils permettent un utilisateur avec relativement peu d’entraînement à soutirer des données importantes qui ont typiquement besoin d’être trouvées par les chimistes entraînés. Le système de workflow de PAINS a fonctionné suffisamment mais les données trouvées étaient problématiques. Le système et un outil en ligne utilisé pour la comparaison des résultats ont signalés des résultats différents, mais les résultats se sont débordés sur les unes les autres. Nous avons trouvés que le système de workflow PubMed alert a très bien fonctionné, car le système pouvait constamment identifier des nouveaux papiers scientifiques. Ces systèmes de workflow sont maintenant implémentés au Consortium Génomique Structurel (SGC) à Toronto. Les deux systèmes de workflow sont disponibles à http://sgc.utoronto.ca/ditommaso.zip. L’implémentation de ces systèmes de workflow montre que le procès est viable et ouvre la voie pour l’implémention des systèmes de workflow plus complexes.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.012
metaresearch head score (Gemma)0.018
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.012
Threshold uncertainty score0.066

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0120.018
Meta-epidemiology (narrow)0.0020.002
Meta-epidemiology (broad)0.0010.003
Bibliometrics0.0020.002
Science and technology studies0.0010.002
Scholarly communication0.0050.005
Open science0.0040.005
Research integrity0.0010.003
Insufficient payload (model declined to judge)0.0110.008

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.016
GPT teacher head0.374
Teacher spread0.358 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2017
Admission routes2
Has abstractyes

Explore more

Same venueJournal of Student Science and TechnologySame topicBiomedical Text Mining and OntologiesFrench-language works237,207