MétaCan
Menu
Back to cohort
Record W3153097092 · doi:10.1136/jmedgenet-2021-107738

Data sharing to improve concordance in variant interpretation across laboratories: results from the Canadian Open Genetics Repository

2021· article· en· W3153097092 on OpenAlexafffundabout
Chloe Mighton, Amanda Smith, Justin Mayers, Robert Tomaszewski, Sherryl A. Taylor, Stacey Hume, Ron Agatep, Elizabeth Spriggs, Harriet Feilotter, Laura Semenuk, Henry Wong, Lorena Lazo de la Vega, Christian R. Marshall, Michelle M. Axford, Talia Silver, George S. Charames, Vanessa Di Gioacchino, Nicholas A. Watkins, William D. Foulkes, Marcos Clavier, Nancy Hamel, George Chong, Ryan E. Lamont, Jillian S. Parboosingh, Aly Karsan, Ian Bosdet, Sean Young, Tracy Tucker, Mohammad R. Akbari, Marsha Speevak, Andrea K. Vaags, Matthew S. Lebo, Jordan Lerner‐Ellis

Bibliographic record

VenueJournal of Medical Genetics · 2021
Typearticle
Languageen
FieldBiochemistry, Genetics and Molecular Biology
TopicGenomics and Rare Diseases
Canadian institutionsWomen's College HospitalBC Cancer AgencyCalgary Laboratory ServicesHospital for Sick ChildrenMcGill University Health CentreMcGill UniversityKingston Health Sciences CentreQueen's UniversityTrillium Health CentreUniversity of ManitobaManitoba HealthUniversity of AlbertaJewish General HospitalSt. Michael's HospitalAgricultural Research Institute of OntarioUniversity of TorontoUniversity of British ColumbiaAlberta Hospital EdmontonUniversity of CalgaryMount Sinai HospitalLunenfeld-Tanenbaum Research Institute
FundersNational Human Genome Research InstituteCanadian Institutes of Health ResearchOntario Genomics Institute
KeywordsConcordanceGenetic variantsComputational biologyMedical geneticsMedicineInterpretation (philosophy)GeneticsComputer scienceBioinformaticsBiologyGenotypeGene

Abstract

fetched live from OpenAlex

BACKGROUND: This study aimed to identify and resolve discordant variant interpretations across clinical molecular genetic laboratories through the Canadian Open Genetics Repository (COGR), an online collaborative effort for variant sharing and interpretation. METHODS: Laboratories uploaded variant data to the Franklin Genoox platform. Reports were issued to each laboratory, summarising variants where conflicting classifications with another laboratory were noted. Laboratories could then reassess variants to resolve discordances. Discordance was calculated using a five-tier model (pathogenic (P), likely pathogenic (LP), variant of uncertain significance (VUS), likely benign (LB), benign (B)), a three-tier model (LP/P are positive, VUS are inconclusive, LB/B are negative) and a two-tier model (LP/P are clinically actionable, VUS/LB/B are not). We compared the COGR classifications to automated classifications generated by Franklin. RESULTS: Twelve laboratories submitted classifications for 44 510 unique variants. 2419 variants (5.4%) were classified by two or more laboratories. From baseline to after reassessment, the number of discordant variants decreased from 833 (34.4% of variants reported by two or more laboratories) to 723 (29.9%) based on the five-tier model, 403 (16.7%) to 279 (11.5%) based on the three-tier model and 77 (3.2%) to 37 (1.5%) based on the two-tier model. Compared with the COGR classification, the automated Franklin classifications had 94.5% sensitivity and 96.6% specificity for identifying actionable (P or LP) variants. CONCLUSIONS: The COGR provides a standardised mechanism for laboratories to identify discordant variant interpretations and reduce discordance in genetic test result delivery. Such quality assurance programmes are important as genetic testing is implemented more widely in clinical care.

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)0.192
metaresearch head score (Gemma)0.451
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch, Open science
Consensus categoriesMetaresearch
DomainCandidate signal: Reproducibility · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.992
Threshold uncertainty score0.997

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.1920.451
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.002
Bibliometrics0.0120.017
Science and technology studies0.0050.003
Scholarly communication0.0090.006
Open science0.0080.016
Research integrity0.0020.003
Insufficient payload (model declined to judge)0.0050.001

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 head0.025
GPT teacher head0.332
Teacher spread0.307 · 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.

Study designObservational
DomainReproducibility
GenreEmpirical

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

Citations36
Published2021
Admission routes3
Has abstractyes

Explore more

Same venueJournal of Medical GeneticsSame topicGenomics and Rare DiseasesFrench-language works237,207