Post-merger R&D integration that turns an acquired biotech's data room into a signed, cited dossier for each asset — before the experts leave
A signed, cited dossier for each asset before the experts leave.
See one case, screen by screen ↓When a pharma company closes on a biotech, it inherits science it has never read: tens of thousands of files across the deal data room, lab notebook and LIMS exports, CRO study reports, batch and stability records and agency correspondence. The integration team has a 100-day plan, and its toxicology, DMPK, CMC, regulatory, IP and clinical owners each need a section they can sign. By our estimate, a function lead spends about five weeks per asset pulling that record together by hand.
The risk sits in what drifted and what was never written down. An IND summary still carries a draft NOAEL that the final report corrected. A polymorph seen in a notebook never reached the CMC report. A human metabolite above 10% of exposure is not covered in a toxicology species. The new sponsor takes on every commitment the former sponsor made to the agency — and the scientists who could explain the record are on retention contracts that end.
Estimated split for one acquired asset, by hand (about 5 weeks) and with the solution (the 6-day target).
Six agents read the data room, cite every fact, find what disagrees and who is leaving; each function owner resolves and signs.
Northwind Bio closed its acquisition of Halden Therapeutics on Aug 18, 2026: four programmes and 38,412 files. Today is Day 50 of the 100-day plan. Here is what happens next in the working solution.
Priya opens a map of the inherited science: four programmes — HLD-217, HLD-330, HLD-405 and HLD-118 — split by function, each tile sized by its files and coloured by how much is now cited in a dossier. 57% is cited and 14,926 facts carry a page. Four authors are leaving, and their workstreams are hatched with the date.
On the HLD-217 DMPK tile: “Only author · 19 days left”.
One click replays the integration from Day 0. On Day 27, Sep 14, the map is at 30% cited and 7,772 facts while the agents read the data room; by today it is 57%. The timeline marks the milestones above and the departures below — Osei left, Solheim leaves, Brandt leaves.
Dr. Ingrid Solheim wrote every one of the 212 M3 bioanalytical assay notebook entries — no one else has signed one — and she can close gaps G-03 and G-06. Priya books Tue Oct 14 with Sam Patel, and a four-question pack goes with the invite: was M3 measured in any rat or dog plasma sample, and where are the raw toxicokinetic files for rat study TX-004?
“Answers are recorded as an interview and linked to the gaps — never as a source record.”
214 files, 1.2 GB, arrived from the Larkfield Preclinical share. The Data Room Classifier sorts them into 41 study reports, 96 data tables, 58 letters and minutes and 19 duplicates, keeping GLP status on the 41 reports. The extractor pulls 612 facts, each with its page. A draft 13-week dog protocol at 5, 15 and 45 mg/kg/day moves gap G-01 forward.
“Halden had already drafted the 13-week dog protocol and the QTc analysis plan — two commitments are further along than we knew.”
In the HLD-217 asset dossier, the toxicology section shows how it was built: 1,412 files labelled for toxicology with drafts and finals told apart, 5 key facts with their pages, 1 contradiction and 2 gaps, 8 citations. The IND Module 2.6.6 summary gives a dog 28-day NOAEL of 50 mg/kg/day, from the draft report of Nov 2, 2023. The final report, Amendment 1, gives 15 mg/kg/day, backed by the pathology peer review.
Only the section owner resolves a contradiction, and both sources are always shown.
On the final value, the exposure margin to the planned 200 mg Phase 2 dose is 1.9×, not 6.2×. Dr. Chen chooses “Use the final report — 15 mg/kg/day”. The IND summary is flagged for correction, the section moves to Ready to sign and HLD-217 readiness goes from 71% to 74%.
“Resolved by Dr. Maya Chen: used the final report value, 15 mg/kg/day; the IND summary is flagged for correction.”
She signs with the meaning of her signature and her password — a section with an open contradiction cannot be signed. The draft section now reads that the dog NOAEL is 15 mg/kg/day from the final report, that the draft value of 50 in the IND summary is flagged for correction in the next annual report, and that the margin to the planned Phase 2 dose is 1.9×. HLD-217 is 78% ready, 3 of 6 sections signed.
Meaning of signature: “I reviewed the cited evidence and approve this section for the integration dossier.”
The Study Reuse Advisor reads the plan, lists 20 candidate mouse models, searches 412 study reports and 9,742 notebook entries, and finds 7 run or tried at Halden. Two are recommended for reuse: MSU crystal peritonitis, where IL-1β fell 48% at 10 mg/kg and 72% at 30 mg/kg, and the LPS/ATP challenge, with an ED50 of 3.1 mg/kg matching human whole-blood potency.
The Study Reuse Advisor cites every prior result and says when a study was stopped or inconclusive — the CAPS knock-in study stopped at week 3.
Five Halden scientists and where each one stands: Dr. Ingrid Solheim leaves Oct 26, Dr. Elias Brandt Nov 14 and Marco Velasquez Dec 19; Hannah Osei has already gone, with 6 of 10 consulting hours left; Dr. Yuki Tanabe is staying. Each person leaving gets a question pack built from the gaps only they can close, and answers are linked back to their gap.
The same acquisition, seen by the people who carry it — the integration lead, the function owners, the program leader and a scientist who is leaving.
Six agents read the data room, cite every fact, find what disagrees and who is leaving; each function owner resolves and signs.
Reads every inherited file in place and labels its record type, asset, study, version and original GLP or GMP status; finds duplicates and drafts.
Pulls studies, findings, PK parameters, lots, stability results and agency commitments into the science registry — each with its document, version and page, and a confidence.
Compares the same fact across versions and sources — draft against final, summary against report, notebook against report — checks the package against agency commitments and guidance, and raises gaps with an owner and an expert who could answer.
Drafts each asset dossier section — toxicology, DMPK, CMC, regulatory, IP, clinical — with every sentence cited and open questions stated plainly; rewrites every sentence that used an old value once a contradiction is resolved.
Finds who authored what, builds a question pack for each departing scientist from the open gaps they wrote or know about, and links their answers back to the gaps.
Answers “has this been done before?” — matches a planned study against the acquired company's reports and notebooks and recommends what to reuse and what is still needed.
Resolves conflicts and signs. The agents propose; a named person decides.
Ask about a study, a lot, a commitment or a person, in plain words. Every answer points to its source.
What contradicts the HLD-217 dog NOAEL?
Two sources disagree. The IND toxicology summary says 50 mg/kg/day, taken from the draft report. The final report, Amendment 1, says 15 mg/kg/day, after a pathology peer review found minimal liver necrosis at 50. On the final value the margin to the planned 200 mg Phase 2 dose is 1.9× instead of 6.2×.
Has Halden already run an NLRP3 mouse model?
Yes — two with clear results. MSU crystal peritonitis cut IL-1β by 72% at 30 mg/kg, and the LPS/ATP challenge gave an ED50 of 3.1 mg/kg that matches human whole-blood potency. A CAPS knock-in study was stopped at week 3.
Who leaves next, and what do they know?
Dr. Ingrid Solheim leaves Oct 26, in 19 days — G-03 on M3 and G-06 on the rat raw data. Dr. Elias Brandt leaves Nov 14 — G-11 on the mouse models. Marco Velasquez leaves Dec 19 — G-04 on Form B and G-08 on lots. Dr. Solheim's 4-question pack is ready.
Which of the inherited science depends on someone who is leaving?
Five workstreams are hatched on the map — their notebooks were written mostly by people who are leaving. The most urgent is HLD-217 DMPK: Dr. Ingrid Solheim wrote every M3 bioanalytical assay notebook, and M3 is above the 10% the agency asked Halden to evaluate.
13 screens from the working solution, on its sample data. Pick one to see it large.
Who signs each dossier section, the rules the agents follow and when to start on a departing scientist are settings.
| Setting | Default | Choose from |
|---|---|---|
| Section sign-off (Toxicology shown) | Dr. Maya Chen | Dr. Maya Chen · Sam Patel · Lena Ortiz · Dana Okafor · Omar Haddad · Priya Raman — one per function |
| Cite the source page for every fact | On · locked | Facts without a page never reach a dossier |
| Keep the original GLP and GMP status | On · locked | Inherited records are read in place, never re-labelled |
| Prefer final and amended reports over drafts | On | On · Off — draft values are shown, but flagged when a final exists |
| Flag values that differ between sources | On | On · Off — raised as a contradiction for the section owner |
| Send low-confidence facts to a person | 0.85 | 0.80 · 0.85 · 0.90 |
| Remind me before a scientist leaves | 30 days before | 14 · 30 · 45 days before |
| Build question packs automatically | On | On · Off — when a gap names someone who is leaving |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample acquisition · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (Verginer et al., “The impact of acquisitions in the biotechnology sector on R&D productivity”, Journal of Business Research, 2025 — about 13.5% higher inventor turnover and about 35% lower citation-weighted patent output; Gulf Coast Consortia toxicology teaching material — about $3 million for a 3-month GLP package in two species). People, companies, programmes and studies named on this page are fictional — characters and sample data in the working solution.
Software for the R&D workstream of an acquisition: turning the acquired company's data room, notebooks, LIMS exports, CRO reports and agency correspondence into a dossier per asset that the acquirer's function owners can sign. Post-Merger R&D Integration does it with six agents that read the record in place and cite every fact to its page.
The Contradiction & Gap Finder compares the same fact across versions and sources — draft against final, summary against report, notebook against report. In the sample acquisition it found the IND toxicology summary still using a draft dog NOAEL of 50 mg/kg/day where the final report says 15, and showed both sources to the head of toxicology.
Workstreams written mostly by scientists who are leaving are hatched on the map with their departure date. The Expert Interview Planner builds a question pack from the open gaps each person can close, and a reminder starts 30 days before they leave by default. Answers are linked back to the gaps.
No. Files are read where they live and nothing is copied or changed. GLP and GMP records keep their original status, and the dossier cites them.
Yes. The Study Reuse Advisor matches a planned study against the acquired company's study reports and notebook entries. In the sample it narrowed 20 candidate mouse models to 2 Halden studies with cited results, and said which study had been stopped.
Yes. The agents propose; only the section owner resolves a contradiction or signs, with the meaning of the signature and a password, and a section with an open contradiction cannot be signed. Expert answers are kept as interviews, never as source records.
The deal data room export, the lab notebook export, LIMS results, CRO study reports, batch and stability records, and regulatory correspondence including the IND — all read only.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your dossier templates, your function owners and a sample of the acquired record — and it goes live once every quality gate has passed. We will walk you through it on your own data first.
Sourced answers from papers, patents and reports in seconds, our own disagreements first.
1-min demoCompetitive Pipeline TrackingEvery rival trial, readout and approval in a sourced brief each Monday
1-min demoPreclinical Study Report AutomationA draft GLP report in hours, every number traced to raw data.
1-min demoWe'll run Post-Merger R&D Integration on a sample of an acquired company's data room.