Subject: Are your organoid sterol dependencies a medium artifact? A 2-week test + genome-wide data

Hi [Name],

We re-analyzed the 2026 Broad (Neiswender et al.) and Sanger (Herranz-Ors et
al.) organoid dependency catalogues genome-wide (DepMap 26Q3, library- and
lineage-adjusted, disjoint held-out-patient confirmed). Headline: 84-86% of
organoid-vs-2D differential dependencies are consistent with medium effects
(reproduced in adherent organoid-medium culture), with a single format
candidate and none at 5% FDR. The strongest
class is sterol/mevalonate synthesis (SQLE delta-Chronos -0.83, ~100%
reproduced without domes), and several sterol/SREBP genes sit on
Herranz-Ors' "organoid-specific core fitness" list (MVD, IDI1, MBTPS1/2,
SCAP). Expression shows the matching sterol-starvation signature, and in
plasma-like medium (Rossiter 2021 screens) SQLE dependency is relieved -
the opposite direction.

If these dependencies reflect LDL starvation in defined medium, LDL rescue
would support LDL limitation as a contributor in these cultured models
(transfer to tumors would remain untested). The cultured-model test takes
~2 weeks without a new screen:

1) LDL add-back in 3-4 of your OPAC/organoid-medium models: SQLE-inhibitor
   and statin dose-response +/- LDL cholesterol (and mevalonate bypass
   control). Our prediction: >=5x IC50 rescue with LDL.
2) Same drugs in matched adherent organoid-medium vs FBS medium.

We pre-agree the falsification criterion: if LDL-replete models keep
SQLE-inhibitor IC50 within 2x in >=3/4 models, our medium-artifact claim
fails and we say so. Either outcome is publishable and directly relevant
to targets now being nominated from these catalogues.

Attached: one-page proposal with design details, the preprint draft, and
the per-gene attribution table (17,525 genes: medium-consistent / format / unresolved
with CIs) so you can check any target you are pursuing. We can rank your
models by predicted effect size if you share ACH/HCM IDs, and we would
analyze any resulting data. Co-authorship on the technical note for the
collaborating lab.

Happy to discuss on a short call.

Best,
[Your name]
[Affiliation]

Attachments: wetlab_proposal_v1.md, manuscript_v2.pdf, medium_attribution_v2_norm.csv (or the full release_v1.zip bundle)
