Verified reads

Research

Public biotech assets, read against the published record -- organized by pathway, ordered by where the deals are. Find the read nearest your asset; then point me at yours. Every claim is two clicks from the filing, the trial, or the paper that backs it.

Sometimes the data is worse than the deck. Sometimes it’s better than the market thinks. I tell you which.

The read

A Phase 3 hazard ratio of 0.16 sounds too good to be true. I read it against the published record -- it holds.

Immunome (IMNM) · gamma-secretase inhibitor · desmoid tumor-- read the report ->

Browse by pathway -- ordered by where the deals are

ADC

1 read

antibody-drug conjugates -- the hottest licensing space of 2025

RNA & oligonucleotide

1 read

three Novartis mega-deals in twelve months

RAS / MAPK & targeted solid tumors

2 reads

the deepest cluster -- where most of the readouts land

AI-designed biologics

1 read

white-hot venture; the diligence is on the asset, not the platform

Congress hubs

EHA 2026 -- heme

The gap between what the deck said and what the data showed, read live on the floor.

See every read on the readout calendar -- by date ->

How to read these reports

Three terms recur -- defined once, so a flag reads as data, not an accusation. An overstatement is a published claim that runs ahead of its own source: common, and usually optimism rather than fraud. A missing-data flag marks something the pitch leans on but that I cannot find in any primary record -- a filing, a trial, a label. A PRR (proportional reporting ratio) measures how often a side effect attaches to one drug versus all others; a high value flags an adverse-event signal worth a closer look. None is a verdict on its own. Treat each as a single data point, and pursue the ones that bear on your decision.

Calibration

What a readout is worth: seven reads written down before ASCO and EHA 2026, scored in the open -- with the misses shown.

A scorecard you cannot lose on is not a scorecard -- read the full card ->

Perspectives

The thinking behind the reads -- long-form arguments from the published record.

Read the essays ->

Every read here was unsolicited -- companies I do not work for. Point me at the one asset you can’t afford to get wrong, and I’ll verify it against the record.

Request a Verified Reador email me directly:chris@bigbio.ai

Start a conversation ->

Reports produced by an AI-assisted diligence pipeline, every finding traceable to a specific database, query, and timestamp. Full terms on the disclaimer.