Skip to content

Talks — ASHG 2026

The primary artifact of this vault: one Markdown file per talk. ASHG 2026 is a hybrid meeting — some talks are tool / pipeline / package debuts, others are clinical-genetics / cohort readouts, others are pure-methods papers. The per-talk schema accommodates all three with one Anchor field.

Status: Template + master index. Per-talk pages will be populated once the program-at-a-glance is public (typically early-September 2026) and the AJHG abstract supplement lands (~September 2026). Pre-meeting stubs flagged *expected, program not yet public* are sourced from ASHG 2025 program shape, lab websites, and cohort-consortium dashboards (All of Us, UK Biobank, gnomAD); titles and assignments will be reconciled with the official program.

Per-talk template

Each talk file (e.g., allan-award-2026.md, aou-cancer-prs.md, gnomad-v5.md) follows this schema. Copy and fill:

# <Talk title — short>

> **Session:** <session name, date, time ET, room> · **Track:** <track / theme> · **Abstract:** <AJHG / abstract #, if assigned>

## At a glance

- **Speaker:** <name, degrees>
- **Affiliation:** <institution, department, lab>
- **Day / session / track:** <Tue Oct 20 / Plenary 1 / Statistical Genetics>
- **Talk title:** <full title as in program>
- **Anchor type:** <tool / package / pipeline | study / cohort / readout | methods paper | award lecture>
- **Anchor:** <package name + Bioconductor/CRAN/PyPI link | study acronym + NCT or cohort | manuscript citation + DOI>
- **Co-authors / consortium:** <if applicable — e.g., All of Us, UK Biobank, GREGoR, gnomAD>
- **Preprint / paper:** <bioRxiv / medRxiv / Nature Genetics / AJHG / HGGA DOI, if accompanying>

## Key result

<One paragraph. Numbers + cohort size + effect size if known. If pre-program, mark
`*expected per <lab page / cohort dashboard / prior-year talk>*` or
`*pre-program speculation*`.>

## Methods / approach

<2–4 sentences. What cohort, what assay, what statistical / computational method,
what comparator. For tool talks: what it consumes / produces, key
methodological idea, language / framework.>

## Why it matters for the corpus

<1–2 sentences. The cancer-genomics relevance specifically — does it inform
hereditary-cancer panel design, cancer-PRS stratification, somatic-mosaicism
detection, or methods used in AACR / ASCO / ASH talks?>

## Cross-links to corpus

- **AACR 2026:** <related session / poster, if applicable>
- **ASCO 2026:** <trial that uses the method / cohort / panel>
- **ASH 2026:** <CHIP / MN-predisposition tie-in>
- **ISMB 2026 / RECOMB 2026:** <methods-paper sibling talk>
- **AGBT 2026:** <upstream platform if this talk applies a new sequencing modality>
- **EuroBioC 2025 tools:** <Bioconductor package used / extended>
- **Other ASHG 2026 talks:** <session-mates, same-cohort, same-method>

## Sources

- AJHG abstract: <DOI / URL or "pending program release">
- Preprint / paper: <bioRxiv / medRxiv / journal DOI>
- Lab page / cohort dashboard: <URL>
- Recording (post-meeting): <ASHG virtual link or YouTube>
- Trade-press recap: <GenomeWeb / Bio-IT URL>

Master talk index

The shortlist below is for stub creation only — all entries are *expected, program not yet public* and will be reconciled with the official ASHG 2026 program when released. Stubs prioritize the cancer-genomics-relevant subset of the program (hereditary cancer, cancer-PRS, somatic mosaicism, clonal hematopoiesis, large-cohort pan-cancer readouts) plus the load-bearing methods / platform talks that we'd want to cite from cancer-talks downstream.

Talk Speaker (expected) Affiliation Day Track Anchor type Anchor (expected) Cancer-genomics relevance
William Allan Award Lecture TBA TBA Plenary (Wed Oct 21) Award award lecture career retrospective Variable — depends on awardee field
Curt Stern Award Lecture TBA TBA Plenary (Thu Oct 22) Award award lecture landmark contribution Variable — Stern often goes to methods or rare-disease investigators with cancer-overlay relevance
Mendel Lecture TBA (international) TBA Plenary (Fri Oct 23) Award award lecture international keynote Variable
Presidential Address / Symposium Susan A. Slaugenhaupt MGH / Harvard Plenary (Tue Oct 20 evening) Society award lecture Communicating Science in an Uncertain World Indirect — public-trust framing
All of Us Research Program update AoU leadership / scientific team NIH / AoU consortium TBD Cohort plenary study / cohort / readout AoU v8 release, expected cancer-cohort overlay High — pan-cancer germline-somatic on >800k participants expected
UK Biobank WGS / proteomics update UKB scientific team UK Biobank TBD Cohort plenary study / cohort / readout 500k WGS, expected cancer-PRS update High — cancer-PRS, hereditary-cancer prevalence
gnomAD v5 release Konrad Karczewski or team Broad Institute TBD Population genetics tool / pipeline gnomAD v5 dataset + browser High — variant-classification backbone for hereditary cancer
GREGoR Consortium update GREGoR PIs NHGRI / GREGoR TBD Mendelian / rare study / cohort / readout unsolved-rare-disease cohort Indirect — cancer-predisposition rare variants
Cancer Genetics platform session multiple Multiple TBD Cancer Genetics mixed hereditary cancer panel updates Core track
Hereditary cancer panel-design / VUS-resolution talks multiple Multiple (incl. Color, Invitae successor labs, academic VUS programs) TBD Cancer Genetics study / cohort ClinVar / VUS reclassification cohorts Core — panel-classification accuracy
Clonal hematopoiesis (CHIP) population-scale talks multiple (Bick / Natarajan / Jaiswal-aligned labs) Multiple TBD Cancer Genetics / Statistical Genetics study / cohort CHIP prevalence + cancer-MN risk in UKB/AoU Core — bridge to ASH MDS/AML talks
Cancer polygenic risk score (PRS) sessions multiple (Mavaddat / Pharoah / Khera-aligned labs) Multiple TBD Statistical Genetics methods paper breast / prostate / colorectal PRS cross-ancestry transferability Core — trial-stratification relevance
Somatic mosaicism in non-cancer phenotypes multiple Multiple TBD Mosaicism methods paper somatic mosaicism callers + cohort applications Methods bleed into cancer ctDNA / clonal-evolution work
Foundation-model variant-effect predictors multiple (DeepMind / EvolutionaryScale / Calico) Multiple TBD Computational / ML methods paper AlphaMissense follow-ons, ESM / EVO genomic models Methods used in AACR variant-classification talks
Long-read clinical sequencing readouts multiple (PacBio / ONT clinical adopters) Multiple TBD Clinical / Diagnostics tool / pipeline clinical WGS programs using HiFi / ONT Methods bleed into AGBT 2026 platform talks
Pan-cancer germline analysis (large cohort) multiple Multiple TBD Cancer Genetics study / cohort germline-pathogenic-variant prevalence across cancers Core — informs panel composition + universal-testing debates
ELSI: AI ethics in genomics multiple Multiple TBD ELSI discussion responsible use of foundation models in clinical genomics Indirect — cancer-AI return-of-results overlap
Trainee Research Excellence Award winners Trainee awardees (announced ~July) Multiple TBD Trainee plenary mixed to be announced Variable but often high-signal

Additional shortlist candidates to stub post-program-release:

  • Cohort updates: MVP (Million Veteran Program) cancer-genomics module; FinnGen R12/R13; GBMI cross-biobank meta-analyses; H3Africa cancer-genomics arm; GenomeAsia 100K cancer-PRS transferability.
  • Methods / tools: DeepVariant / GLnexus updates; CADD v2; PRS-CSx / SBayesRC cross-ancestry methods; mosaic-variant callers (MosaicHunter, deepMosaic); pangenome / HPRC clinical applications.
  • Pre-meeting workshops to flag: statistical-genetics primers; clinical-sequencing pipelines; PRS construction hands-on; foundation-models-for-genomics tutorials.

Cross-cutting themes (for themes.md, deferred until post-meeting)

To synthesize across talks/ once populated:

  1. Cancer-PRS clinical utility moment. Cross-ancestry PRS transferability + biobank-scale validation may finally cross the threshold for trial-stratification adoption. Cross-link to AACR 2026 PRS sessions and ASCO 2026 breast / prostate biomarker-eligible trials.
  2. gnomAD v5 + ClinVar baseline. New population reference may re-classify a meaningful fraction of variants in hereditary-cancer panels. Cross-link to AACR 2026 variant-classification posters.
  3. Long-read clinical sequencing inflection. PacBio HiFi + ONT R10 quality at clinical scale — cross-link to AGBT 2026 platform talks.
  4. Foundation models for variant effect. AlphaMissense follow-ons + protein-language-model + DNA-LM successors. Cross-link to AACR 2026 ML-bio sessions and ISMB 2026.
  5. Somatic mosaicism beyond cancer. Methods sharing with cancer-ctDNA / clonal-evolution methodology.
  6. CHIP / CCUS at population scale. Bridge to ASH 2026 MDS / AML clinical talks.
  7. AoU + UK Biobank coordinated cancer readouts. Two cohorts both with ~mature cancer-genomics modules; pan-cancer germline-somatic at population scale is now feasible.