Skip to content

RECOMB 2026

30th Annual International Conference on Research in Computational Molecular Biology — Thessaloniki Concert Hall, Thessaloniki, Greece · May 26–29, 2026. In-person. Satellite workshops + H.bioinfo'17 run May 24–25.

Status: Scaffold landed pre-meeting (~15 days out as of May 11, 2026). Program is fixed (full session-by-session schedule public), all 4 keynote speakers announced, and the complete 65-paper accepted-list is already public — so unlike ISMB 2026 (where the proceedings list is still embargoed in May), RECOMB 2026 tool-page extraction can be scaffolded before the meeting from titles + bioRxiv preprints, then enriched with talk-day metadata + slide DOIs immediately post-meeting.

One factual correction up front: the task brief framed RECOMB 2026 as having "concluded last month." It hasn't — the conference runs May 26–29, 2026, which is ~2 weeks in the future as of today (May 11, 2026). The build-out below treats the meeting as upcoming. The accepted-papers list and program schedule are already public; what's missing is talk delivery, best-paper announcements, and slide/recording deposits.

Why this is in the vault

  • Pure methods/algorithms venue. RECOMB is the premier theoretical computational-biology conference — smaller and tighter than ISMB (65 accepted papers vs. ISMB's ~600 talks), peer-reviewed proceedings, no industry/translational/community tracks to wade through. Every accepted paper is a methods or algorithms contribution with a code repo. Highest signal-to-noise ratio of any vault we maintain.
  • Foundational-method pipeline upstream of AACR. RECOMB papers tend to be the algorithmic substrate (alignment, phylogeny, single-cell statistics, foundation-model architectures, privacy-preserving inference) that later shows up productized at ISMB, applied at AACR, or wrapped at Nextflow Summit / BOSC. Watching RECOMB lets us anticipate which tools will surface at AACR 2027 / EuroBioC 2026 / GBCC 2026.
  • AACR axis fit — narrower but deep. The 65-paper program breaks down into 6 tight bins: single-cell (CAMP, DiCoLo, scDesignPop, MOSAIC, scProfiterole, GeoAdvAE, Gene-First, CycleGRN, DeltaNMF, error-correction, Nullstrap-DE, Information-Geometry), cancer evolution (Arborist, Cycle-Extractor for ecDNA, POTTR, subclonal tree alignment), spatial omics (MOSAICField, DIME, LineageMap), protein/ML (ProtFlow, MutPred-PPI, MSA-Transformer hybrids, foundation-model adapters, synthetic-lethality predictors), statistical/privacy genetics (Private PRS leakage, GOPHER, biobank G×E, pangenome privacy), and sequence/phylogeny algorithms (Cuttlefish 3, PBWT/SLAB, minimizers, quartet trees, STELAR-X). Bins 1–4 are directly relevant to AACR; bin 5 is relevant to AACR-axis privacy/PRS work; bin 6 is upstream infrastructure.
  • Cross-vault people overlap. Bonnie Berger (MIT, steering chair) and Anna Sappington (Berger lab) appear on 2 RECOMB papers and overlap with ISMB MLCSB. Olga Troyanskaya (Princeton, ISMB 2026 keynote on Mon Jul 13) co-authors a RECOMB paper here (Aggarwal/Sokolova/Troyanskaya on multi-modal tissue-aware GNN). Sara Mostafavi (RECOMB keynote Fri May 29) co-authors paper #19 ("Deep genomic models of allele-specific measurements"). Ben Raphael (Princeton) is on 3 papers. Rob Patro is on Cuttlefish 3 (paper #49) — same person whose tximeta/fishpond surfaces at GBCC 2025 and ISMB 2026 (BOSC). Jingyi Jessica Li (UCLA) authors 3 papers (Information-Geometry, Nullstrap-DE, scDesignPop) — high-overlap with ISMB MLCSB.

What we have / will have to work with

Source Coverage Notes
Accepted-papers page full 65-paper list, titles + author lists, no abstracts recomb.org/recomb2026/accepted_papers.html — extracted in program-notes.md
Program schedule page 14 sessions across 4 days, 4 keynotes, paper-to-session mapping recomb.org/recomb2026/program.html — extracted in program-notes.md
Keynotes page 4 distinguished speakers; talk titles not yet public recomb.org/recomb2026/keynotes.html — speaker images only; titles likely posted day-of or via Twitter/Bluesky
Committees page steering, program chair, organizing, posters, fellowships recomb.org/recomb2026/committees.html
Call for Papers proceedings + dual-submission rules; no Springer LNBI this year recomb.org/recomb2026/call_for_papers.html
Proceedings (no LNBI) RECOMB 2026 breaks the LNBI tradition — papers route to partnering journals instead Partnering journals: Cell Systems, Genome Research, Journal of Computational Biology (default special-issue venue). Authors of selected papers may be invited to submit expanded versions to Cell Systems or Genome Research for expedited review; the remainder land in a JCB special issue.
bioRxiv / arXiv preprints most RECOMB papers have a public preprint recomb.org/proceedings/proceedings/2030-2026/2026/ is the running proceedings hub and includes preprint links per paper
Satellite-workshop sites 7 satellites Sun–Mon May 24–25 RECOMB-Seq, RECOMB-CCB, RECOMB-CG, RECOMB-Genetics, RECOMB-Privacy, RECOMB-Arch, RECOMB-RSG. H.bioinfo'17 also May 25. Per-satellite programs at recomb-seq.github.io, recomb-cg.org, etc. — populated separately.
Slide deposits ad-hoc; no central RECOMB Zenodo community Expect lab-site or speaker-Bluesky deposits. Berger lab + Mirarab lab + Klau lab historically deposit slides; many others don't.
Recordings RECOMB main program traditionally not recorded Recheck post-meeting; some sessions occasionally posted to the RECOMB YouTube channel ~6–10 weeks late
Press / social #RECOMB2026 on Bluesky + X useful for talk reactions and best-paper announcements (named on stage at the closing session, Fri May 29)

Program shape

RECOMB 2026 runs Tue May 26 through Fri May 29, single-track main program (no parallel sessions during the conference proper). Each day has 3–4 paper sessions of 3–6 talks each, plus 1 keynote (Tue/Wed/Thu/Fri). Total: 14 paper sessions, 65 contributed talks, 4 keynotes. Satellites + H.bioinfo'17 run on the two days before (Sun May 24 + Mon May 25) as parallel side events; same venue but different rooms.

Counts:

  • 4 Distinguished Keynotes: Paul Medvedev (Tue), Gene Myers (Wed), Alexandros Stamatakis (Thu), Sara Mostafavi (Fri)
  • 14 paper sessions organized by topic: Sequence I/II, Protein/ML I/II, Single-Cell I/II/III, Multi-Omics, Small Molecules, Transcriptomics, Phylogeny I/II, Cancer & Evolution, Statistical Genetics
  • 65 peer-reviewed accepted papers (PC Chair: Rayan Chikhi, Institut Pasteur; PC ~100 members)
  • 7 satellite workshops May 24–25 (RECOMB-Seq, -CCB, -CG, -Genetics, -Privacy, -Arch, -RSG) + H.bioinfo'17 (Hellenic Bioinformatics Society) May 25
  • Posters — multiple sessions; not yet enumerated by the public page (Posters Chair: Kaiyuan Zhu)
  • No industry track, no tutorials, no parallel "special tracks" — single-track main program is the entire conference. This is the major shape difference from ISMB.

Organization

conferences/recomb-2026/
├── index.md             # this page
├── program-notes.md     # internal working file — full 65-paper list + day-by-day grid + tool shortlist
├── themes.md            # cross-session synthesis (deferred until post-meeting)
├── digest/              # day-by-day recap (deferred until post-meeting)
│   ├── day-1-tuesday.md     # Medvedev keynote + Sequence I + Protein/ML I + Single-Cell I
│   ├── day-2-wednesday.md   # Myers keynote + Sequence II + Multi-Omics + Small Molecules + Transcriptomics
│   ├── day-3-thursday.md    # Stamatakis keynote + Phylogeny I/II + Single-Cell II + Cancer & Evolution
│   └── day-4-friday.md      # Mostafavi keynote + Protein/ML II + Statistical Genetics + Single-Cell III + closing/best-paper
├── keynotes/            # one entry per keynote (deferred)
├── satellites/          # per-satellite-workshop entries (deferred; populated from per-satellite sites)
│   ├── recomb-seq.md
│   ├── recomb-ccb.md
│   ├── recomb-cg.md
│   ├── recomb-genetics.md
│   ├── recomb-privacy.md
│   ├── recomb-arch.md
│   └── recomb-rsg.md
├── topics/              # cross-cutting bins (deferred)
│   ├── single-cell-methods/      # 12+ papers across Sessions 3, 9, 14
│   ├── cancer-evolution/         # Session 11 (Arborist, ecDNA, POTTR, subclonal trees)
│   ├── spatial-omics/            # MOSAICField, DIME, LineageMap (Sessions 5+9)
│   ├── protein-ml/               # Sessions 2 + 12
│   ├── privacy-statgen/          # Session 13 + PangenomePrivacy + GOPHER + PRS-leakage
│   └── sequence-phylogeny/       # Sessions 1, 4, 8, 10 — upstream-infra bin
├── tools/               # PRIMARY artifact — one entry per package / method
│   └── index.md
└── sessions/            # full per-session abstract index (post-meeting)
    └── index.md

The tools/ directory is the load-bearing artifact, same shape as EuroBioC2025 / GBCC2025 / ISMB 2026. RECOMB-specific additions to the per-tool template (see tools/index.md): a Proceedings row pointing to the partnering-journal DOI (Cell Systems / Genome Research / JCB) when known, a Session field naming the RECOMB session (e.g., "Session 11: Cancer & Evolution"), and a Preprint row for the bioRxiv/arXiv link that exists for most papers.

What's different from EuroBioC / GBCC / ISMB

  • Pure-methods filter. EuroBioC = R/Bioconductor community workshop. GBCC = Galaxy + Bioconductor joint, applied focus. ISMB = 22-track everything-soup with industry + tutorials + BOSC + COSI politics. RECOMB = algorithms only, single track, peer-reviewed proceedings. Every accepted paper is a tool/method/theory contribution. Tool-page coverage ratio is the highest of any vault: 65 papers → likely ~50–60 tool pages.
  • No Springer LNBI this year. Previous RECOMB years published proceedings as a single Springer Lecture Notes in Bioinformatics (LNBI) volume. RECOMB 2026 breaks this tradition: papers route to Cell Systems, Genome Research, or Journal of Computational Biology (default). This raises the citation profile per paper but means DOI lookup is per-paper, not per-volume — slows down tool-page enrichment vs. an LNBI year.
  • Smaller scale, single track. ~250–400 attendees (vs. ISMB ~3000, EuroBioC ~250, GBCC ~300). No parallel sessions during the main program — so unlike ISMB, every attendee sees every talk. Tool-page extraction doesn't need a parallel-track triage step.
  • No tutorials, no BOSC equivalent. RECOMB-CCB and RECOMB-RSG are the closest analogues to applied tracks, but they're separate satellites on May 24–25 — not nested inside the main program. Bioconductor / scverse / Galaxy ecosystem packages rarely debut at RECOMB; when they do, it's the underlying algorithm paper, not the package release.
  • Heavy phylogeny / sequence-algorithm load. ~20 of 65 papers (≈30%) are in Sessions 1, 4, 8, 10 — pure sequence algorithms, phylogenetic inference, pangenome indexing, minimizers, PBWT, de Bruijn graphs. This bin has minimal AACR axis fit and is treated as upstream infrastructure (one paragraph per paper in topics/sequence-phylogeny/, not full tool pages unless the artifact is a maintained tool with a release — e.g., Cuttlefish 3, MaxGeomHash).
  • Closing-day best-paper announcement. RECOMB traditionally announces best-paper and best-student-paper awards at the closing session (Fri May 29 PM). Updating the vault to flag winners is a same-day post-meeting task.

Sources

Next step

  • Now (pre-meeting, May 11–26): the accepted-paper list is already public, so scaffold likely tool pages from titles + bioRxiv preprints. The program-notes.md shortlist below identifies the ~12–15 highest-priority entries. No need to wait.
  • During the meeting (May 26–29): real-time capture of (a) keynote talk titles when announced from stage, (b) best-paper / best-student-paper award winners at the closing session Fri May 29, (c) slide deposits posted to #RECOMB2026 social, (d) any preprint-to-journal redirects (Cell Systems / Genome Research / JCB acceptances announced at the closing).
  • Satellite-workshop pass (May 24–25): scaffold the 7 satellites + H.bioinfo'17 in satellites/. Each gets its own page with workshop-specific accepted-papers list (separate from main RECOMB).
  • Bulk tool-page extraction (post-meeting, May 30+): ~50–60 candidate entries spanning the 65 papers. Existing-in-corpus tools (Cuttlefish, Voyager-adjacent spatial work, SpatialData spillovers, Tidyomics-adjacent statistical-genetics work) get an "RECOMB 2026 update" section appended; new RECOMB-debuting tools get fresh entries.
  • Partnering-journal DOI sweep (Jul–Sep 2026): the partnering-journal track means each accepted paper acquires a final DOI on a per-paper schedule, not a single LNBI volume drop. Plan a re-sweep in Jul / Sep to backfill Cell Systems / Genome Research / JCB DOIs into the tool-page Proceedings rows.