Abstract 5287: Heterogeneity of mutation calling and annotation: a survey of cancer next-generation sequencing initiatives by the Global Alliance for Genomics and Health (GA4GH)
Bibliographic record
Abstract
Abstract Although genomic data sharing is widely endorsed, practical barriers exist. For example, annotation and calling of mutations vary significantly with the method used. This poses clear challenges in data harmonization and sharing. In this setting, GA4GH conducted a survey of Cancer Next Generation Sequencing (NGS) initiatives globally to chart the technical implementation of genomic programs. A total of 59 of 108 invited initiatives responded (response rate = 55%) via a web-based survey. In total, 63% of programs share their data, and 10% are partially sharing or planning to share. Most initiatives were North American (33%) or European (28%) based. Of the 59 respondents, 51 responded to queries on technical aspects of their NGS program (Table 1). For diagnostic application, 67% employ a small panel (<50 genes), 55% a medium panel (50-250 genes), 45% a large panel (251-1000 genes); only 22% indicated they (also) use whole exome sequencing (WES). Diagnostic programs tend to favor panels and deep sequencing, while research programs favor WES/WGS. Overall, mutation and copy number calls were stored centrally in a single project database (96% and 92% respectively), with lower rates for raw data (BAM files, 86%) and histological data (75%). Somatic mutations were identified primarily via GATK (57%), Samtools (49%), Varscan (47%), MuTect (40%); all but 7 initiatives used combinations of these tools. Mutational variants were annotated by Cosmic (73%), PolyPhen (66%), and dbSnp (64%). Germline samples were used as control in only 62% of initiatives. In conclusion, the majority of initiatives use an ensemble of tools for calling and annotating mutational variants. Harmonization efforts on gene panel composition and the standardization of tools (eg, methods, application program interfaces (APIs)) are urgently needed to prevent continued generation of isolated data silos that hamper NGS-enabled advances in precision medicine. Primary purpose of test% Diagnostic (n = 9)% Research (n = 22)% Diagnostic & Research (n = 20)OrganizationCentralized testing22%64%30%Unique sample identifiers89%77%70%Certification ISO/CLIA/NE88%45%65%Initiative > 1000 Patients33%36%35%SequencingSequencing depth 50-250x0%55%80%Sequencing depth >250x100%45%20%Panel sequencing89%68%60%Whole Exome/Genome Sequencing22%77%65%SamplesFresh frozen (FF)11%23%15%Formalin Fixed Paraffin Embedded (FFPE)44%27%20%FFPE or FF44%50%65%Germline as control22%68%65% Citation Format: Daniel J. Vis, Jeremy Lewin, Lillian Siu, Rachel Liao, Jean Claude Zenklusen, Fabien Calvo, Edit Szepessy, Ana Vivancos, Valtteri Wirta, Subha Madhavan, Keunchil Park, Daniel Tan, Janessa Laskin, Melissa Brammer, Emmanuel Dias-Neto, Anthony Tolcher, Thomas J. Hudson, Charles Sawyers, Mark Lawler, Emile E. Voest. Heterogeneity of mutation calling and annotation: a survey of cancer next-generation sequencing initiatives by the Global Alliance for Genomics and Health (GA4GH). [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5287.
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 | 0.037 | 0.045 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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".