Abstract 449: A standard operating procedure for the curation of gene fusions
Bibliographic record
Abstract
Abstract Despite the well-established role of recurrent gene fusions as oncogenic drivers, current practices for characterizing and interpreting gene fusion events in clinical testing and in biomedical literature are inconsistent. From the conceptual definition of gene fusions to the salient elements that characterize these alterations, a lack of community-driven standards for the curation of gene fusions has resulted in a disparate landscape of fusion representations and supporting tools. Consequently, the evidence-based clinical evaluation of gene fusions requires extensive expert review for accurate interpretation of observed gene fusions with respect to putative evidence from biomedical literature. Furthermore, the lack of these standards inhibits the interoperability of tools, resources, and pipelines - impeding data sharing and downstream utility.To address these challenges, a cross-consortia initiative between the Variant Interpretation for Cancer Consortium and ClinGen was formed to develop a standard operating procedure (SOP) for the curation of gene fusions. The SOP is under development by an international and diverse set of experts in the representation, detection, and clinical interpretation of gene fusions. Participating stakeholders across academic, government, and industry sectors showcased challenges and solutions, and participated in community surveys and discussions to define and develop the SOP for this diverse class of alterations.An initial result of this effort was the precise molecular definition of genomic events and features constituting gene fusions. We distinguish these from similar but distinct classes of structural alterations through clinically-relevant examples. Next, we discuss our findings on community practices around the description and evaluation of gene fusions. We provide our recommendations for characterization and representation of gene fusions from these practices, and compare these recommendations to existing variant representation standards and formats (e.g. HGVS variant nomenclature). We also discuss the concurrent application of formats for standardized human- and machine-readable representations of gene fusion events.We conclude with discussion of the salient elements to enable rapid, scalable, and consistent evaluation of fusions curated from the biomedical literature. Recommendations are provided for the standardized capture of these elements to enable both intuitive and precise characterization of this diverse class of alterations in clinical reporting and literature. In summary, we provide a clinical-practice driven framework and nomenclature for gene fusions, including recommendations for human readability, computational precision, and data integrity within the SOP. This work is a substantial advancement towards standardized communication, investigation, and sharing of gene fusion data across clinical and research domains and specialties. Citation Format: Alex H. Wagner, Ioannis S. Vlachos, Dmitriy Sonkin, Panieh Terraf, Chimene Kesserwan, Andrea Sboner, Thomas Coard, Christian Reich, Deborah I. Ritter, Peter Horak, Ying S. Zou, Anna Tanska, Aaron M. Berlin, Anna Lu, Daniel Cameron, Heather E. Williams, Wan-Hsin Lin, Gokce Toruner, Arpad Danos, Jason Saliba, Huiling Xu, Xinjie Xu, Georgina Ryland, Michele Ceccarelli, Liying Zhang, Sarah Rapisardo, Catherine Rehder, Xuelu Liu, Aparna Pallavajjala, Nicole Park, Laveniya Satgunaseelan, Kristy Lee, Jie Liu, Obi Griffith, Robert R. Freimuth, Albrecht Stenzinger, Linda B. Baughn, Michael Baudis, Jennifer Lee, Marilyn Li, Angshumoy Roy, Gordana Raca. A standard operating procedure for the curation of gene fusions [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 449.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.058 | 0.138 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.010 | 0.005 |
| Science and technology studies | 0.003 | 0.003 |
| Scholarly communication | 0.007 | 0.005 |
| Open science | 0.004 | 0.009 |
| Research integrity | 0.004 | 0.006 |
| Insufficient payload (model declined to judge) | 0.045 | 0.053 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".