Per-program dossiers — AGBT Precision Health 2026¶
One page per clinical study, biobank initiative, real-world implementation program, or precision-therapeutic case presented at AGBT-PM 2026. Because AGBT-PM abstracts are closed (attendee-only), the per-program template — not per-talk — is the natural unit, with each dossier reconstructed from public coverage of the speaker's program rather than the talk itself.
Status: Pre-meeting scaffold (T-4 months as of May 11, 2026). The index below will be populated after the September 14–16, 2026 meeting from confirmed speaker affiliations + their public program output. Pre-meeting, this page documents the template and the anticipated dossier candidates based on the 15 confirmed speakers.
Per-program template¶
Each dossier follows this shape:
# <Program / Trial / Initiative name> — <PI / institution>
- Presenter / session / date at AGBT-PM 2026
- Program type: rare-disease dx | MRD trial | population biobank | gene-therapy n=1 | implementation | pharmacogenomics
- Cohort size / current enrollment / endpoints
- Key reported findings (with source citation — vendor PR, society PR, preprint, peer-reviewed paper, podcast)
- Platform / assay used (cross-link to AGBT 2026 launch dossier if applicable)
- Clinical-implementation status: research-only | LDT | FDA-cleared | reimbursed | guideline-recommended
- Reception / competitive context (Guardant vs Natera vs Personalis vs Foundation, etc.)
- Cross-links: AACR / ASCO / ESMO / SITC / ASHG sibling read-outs; AGBT GM platform vendor
Every clinical claim ends with a citation. Distinguish interim vs mature read-outs. Note n explicitly.
Anticipated dossier candidates (pre-meeting forecast)¶
Built from the 15 confirmed speakers' affiliations and recent public output. Not yet populated — these are scaffold slots to be filled after the meeting.
Rare-disease genomic diagnostics¶
| Anticipated program | PI / institution | Source for forecast |
|---|---|---|
| GREGoR Consortium — diagnostic-odyssey rare-disease cohort | Heidi Rehm (Broad), with multi-site network | NIH GREGoR |
| BabySeq / Rapid-WGS in NICU | Monica Wojcik (Boston Children's) | Recurring AGBT-PM thread; multiple peer-reviewed cohorts |
| Penn Rare Disease Program | Theodore Drivas (Penn) | Recurring Penn medical-genomics implementation |
| Long-read clinical sequencing for rare disease | Andrew Stergachis (UW) | Stergachis Lab — Fiber-seq / phased clinical assemblies |
| Broad clinical sequencing platform | Niall Lennon (Broad) | Broad Clinical Labs throughput / case mix |
AI in clinical genomics¶
| Anticipated program | PI / institution | Source for forecast |
|---|---|---|
| DL for variant interpretation / functional genomics | Alexis Battle (JHU) | Recurring AGBT-PM speaker on QTL / DL methods |
| Foundation models for genomic medicine | Sara Mostafavi (UW / Allen) | Recent enformer / genomic-LLM work |
| eMERGE network — clinical-decision support | Marylyn Ritchie (MUSC) | eMERGE phase IV implementation reports |
Precision therapeutics & gene editing¶
| Anticipated program | PI / institution | Source for forecast |
|---|---|---|
| In-vivo CRISPR for monogenic disease | Kiran Musunuru (Penn) | VERVE-101 / individualized base-editing for FH program |
| Custom CRISPR therapeutics platform | Fyodor Urnov (IGI Berkeley) | n=1 / ultra-rare editing therapeutic pipeline |
Population genomics & biobanks¶
| Anticipated program | PI / institution | Source for forecast |
|---|---|---|
| Regeneron Genetics Center — exome+ | Gonçalo Abecasis (Regeneron) | Recurring; latest WES/WGS cohort updates |
| UK Genomic Medicine Service / Oxford | Jenny Taylor (Oxford) | Genomics England 100K → NHS GMS transition |
| NIH All of Us research workbench | (TBD — All of Us workshop in past years) | All of Us AGBT-PM 2025 workshop |
Newborn / prenatal precision sequencing¶
| Anticipated program | PI / institution | Source for forecast |
|---|---|---|
| NICHD prenatal sequencing | Diana Bianchi (NICHD) | NICHD director, prenatal-sequencing portfolio |
| Newborn WGS pilots | Monica Wojcik (Boston Children's) | BabySeq2, NewbornsInSC, GUARDIAN comparisons |
Vendor / industry tracks (expected, not yet sourced to confirmed speaker)¶
- Guardant — Reveal MRD + Shield colorectal screening real-world performance (likely vendor-track)
- Natera — Signatera MRD across indications (likely vendor-track)
- Personalis — NeXT Personal Dx ultra-sensitive MRD (likely vendor-track)
- Foundation Medicine / Tempus / Caris — comprehensive genomic profiling implementation
- Illumina / Element / Roche / Ultima — clinical-throughput claims tied to AGBT GM 2026 launches
Confirmed dossiers¶
None yet — meeting is T-4 months out. Per-program dossiers will be added during and after the meeting from public coverage of each confirmed speaker's program.
How to add a dossier¶
- Create
trials/<pi-or-program-slug>.md(e.g.trials/wojcik-babyseq.md,trials/regeneron-genetics-center.md). - Follow the per-program template above.
- Every clinical figure (cohort n, sensitivity/specificity, PPV/NPV, OS/PFS, diagnostic yield) must end with a citation to a peer-reviewed paper, preprint, society press release, or vendor PR — never reconstruct from a tweet alone.
- Cross-link to:
- Upstream platform in
../../agbt-2026/launches/if the program uses an AGBT-GM-launched instrument. - Sibling read-outs at AACR 2026, ASCO 2026, ESMO 2026, SITC 2026, ASHG 2026.
- Tool / pipeline dossiers in EuroBioC 2025 / GBCC 2025 / Nextflow 2026 when the program is built on a publicly named workflow.
- Flag any vendor figure that lacks third-party validation as (vendor PR — no independent validation).
Caveats specific to AGBT-PM dossiers¶
- Clinical implementation programs evolve faster than the conference cycle. A "real-world implementation" presented in September 2026 reflects data cut some months earlier. Note the data-cut date when known.
- Patient privacy. Storytelling sessions and rare-disease case reports often include patient-family voices; do not reconstruct any patient-identifying detail beyond what the program has publicly released.
- Vendor MRD claims are routinely PR-distorted. Each MRD vendor (Natera, Guardant, Personalis) has different limit-of-detection, longitudinal cadence, and indication-specific evidence. Compare to a common ASCO/ESMO endpoint (DFS, MRD-detection lead time) rather than vendor's preferred metric.
- GREGoR diagnostic-yield numbers vary substantially by recruitment criteria (prior-WES-negative vs unselected, prior-WGS-negative vs WES-naive). Always include cohort selection in the dossier.