Biology Atlas

Map / The cascade

step · intermediate · partly written

Hit discovery

One step of the screening cascade walk-through, from disease to clinic.

What is done

Find compounds that move the target's readout. The sources show two routes. One tests a physical library in a plate assay, as the Broad STK33 programme did with a library of 321,811 compounds in 1536-well format. The other has computational groups predict binders and then tests the predictions: in CACHE challenge 1, 23 teams each predicted up to 100 compounds for the WDR domain of LRRK2, a protein with no known ligands, and 1,955 molecules were tested by surface plasmon resonance. A "hit" at this step means a signal change at one concentration, not a drug-like molecule. 123

Reported, not yet replicated

Costs

The public prices are for the screen itself. One academic core's FY25 internal rates are $1.00 per well for a small-molecule 384-well screen with a typical $7,500 setup fee, compounds and the assay not included. A 2020 recharge sheet from another academic core prices a 25,000-compound primary biochemical screen at $7,041 internal and $13,377.90 external, reagents supplied by the lab. No contract-lab price and no whole-campaign cost was found; buying a library and developing the assay are outside these figures, and academic recharge rates are not market prices. 45

Reported, not yet replicated

Proves

Only that, under the screen's conditions, a compound changed the readout. In the STK33 programme the single-concentration kinase screen record lists 321,908 substances and 235 actives, and an 8-concentration retest of 224 substances called 174 active. 671

Reported, not yet replicated

Cannot prove

That the compound binds the target, acts through it, is selective, or does anything in a cell. The STK33 probe ML281 had an enzyme IC50 of 14 nM yet did not meaningfully reduce cell viability up to 10 uM, and a separate study concluded (abstract read) that STK33 kinase inhibition does not look like a promising anti-KRAS strategy. Readout chemistry also produces false hits: in one profiling study firefly luciferase was inhibited by 12% of 360,864 compounds (abstract only), and in 3,048 Lilly biochemical assays compounds matching PAINS alerts were active 13.8% of the time against 9.6% overall. 18910

Reported, not yet replicated

Assays

Regulators require

No regulation read for this atlas says how hits or leads must be found. The first document the regulator requires is the investigational new drug application, whose pharmacology and toxicology section is built from laboratory-animal or in vitro studies on which the sponsor concludes it is reasonably safe to begin trials (21 CFR 312.23(a)(8)). Discovery data reaches the regulator only as the sponsor's own supporting evidence. This is our reading of the rule, not a statement in it. 11

Our inference

ML role

Does: predicted hit lists can be scored against a measured binding assay. In CACHE challenge 1, 73 of 1,955 tested molecules bound the LRRK2 WDR domain in SPR with a dissociation constant under 150 uM, a prospective baseline for a computational hit list (abstract and an NSF record read; full text not opened). Models trained on Tox21 data flag luciferase inhibitors and autofluorescent compounds with accuracies of about 80%, so likely artefacts can be filtered before a counter-screen. Cannot: a binder in SPR may be inert, and a prediction is not a measurement of function. 2312

Reported, not yet replicated

Reports

Scenarios

Modality notes

Antisense: the hit search is a walk along a sequence

For an antisense drug the target is chosen by what the RNA looks like in sequence, and selectivity becomes sequence uniqueness, so many sequences are screened along the target (tiling) rather than a library of pocket-binders. The modality record calls this an inference from its sources. 131415

See: Antisense oligonucleotide (ASO)

Our inference

siRNA: which sequence is cut and survives

The design question moves from what binds the protein to which sequence is cut efficiently and which chemical modifications make it stable. 161718

See: siRNA

Our inference

Still open here

Linked from

Sources

  1. Weiwer et al.. Weiwer et al. 2012: the STK33 probe ML281 (PMC full text). PubMed Central (2012). paper; opened 2026-11-17 all records citing this
  2. Li et al.. CACHE Challenge #1: LRRK2 WDR domain hit finding (Li et al., 2024). Journal of Chemical Information and Modeling (abstract via Europe PMC) (2024). paper; opened 2026-11-17 all records citing this
  3. NSF PAR record 10554378 for the CACHE Challenge #1 paper. NSF Public Access Repository (2024). web, not peer reviewed; opened 2026-11-17 all records citing this
  4. Ohio State University High Throughput Screening Core: cost structure (FY25). Ohio State University (2025). organisation-page, not peer reviewed; opened 2026-11-17 all records citing this
  5. MSU Assay Development and Drug Repurposing Core recharge rates (file dated 2020-05-27). Michigan State University (2020). report, not peer reviewed; opened 2026-11-17 all records citing this
  6. PubChem BioAssay AID 2661: STK33 primary screen (ADP-Glo). PubChem. database, not peer reviewed; opened 2026-11-17 all records citing this
  7. PubChem BioAssay AID 2821: STK33 inhibitor retest, 8-point dose response. PubChem (2010). database, not peer reviewed; opened 2026-11-17 all records citing this
  8. Luo, T., et al.. Luo et al. 2012: STK33 kinase inhibitor BRD-8899 in KRAS-dependent cells, PNAS. PubMed Central (2012). paper; opened 2026-11-17 all records citing this
  9. Thorne N, Shen M, Lea WA, et al.. Firefly luciferase in chemical biology: a compendium of inhibitors, mechanistic evaluation of chemotypes, and suggested use as a reporter. Chemistry & Biology 19(8):1060-1072 (2012). paper; opened 2026-11-17 all records citing this
  10. Lilly 2018 analysis of PAINS in biochemical assays (PMC full text). PubMed Central (2018). paper; opened 2026-11-17 all records citing this
  11. US Food and Drug Administration. IND content and format (21 CFR 312.23). Electronic Code of Federal Regulations, Title 21, Part 312 (2026). regulatory, not peer reviewed; opened 2026-11-17 all records citing this
  12. Borrel A, Huang R, Sakamuru S, et al.. High-throughput screening to predict chemical-assay interference. Scientific Reports 10:3986 (2020). paper; opened 2026-11-17 all records citing this
  13. Egli, M., Manoharan, M.. Chemistry, structure and function of approved oligonucleotide therapeutics. Nucleic Acids Research (2023). review; opened 2026-11-17 all records citing this
  14. U.S. Food and Drug Administration (label text supplied by the sponsor). SPINRAZA (nusinersen) injection, for intrathecal use: US prescribing information (initial US approval 2016, revised 12/2016). FDA Drugs@FDA label archive (2016). regulatory, not peer reviewed; opened 2026-11-17 all records citing this
  15. Hua, Y., et al.. Antisense masking of an hnRNP A1/A2 intronic splicing silencer corrects SMN2 splicing in transgenic mice. American Journal of Human Genetics (2008). paper; opened 2026-11-17 all records citing this
  16. Foster DJ, et al.. Advanced siRNA designs further improve in vivo performance of GalNAc-siRNA conjugates. Mol Ther (2018). paper; opened 2026-11-17 all records citing this
  17. Adams D, et al.. Patisiran, an RNAi Therapeutic, for Hereditary Transthyretin Amyloidosis. N Engl J Med (2018). paper; opened 2026-11-17 all records citing this
  18. Judge DP, et al.. Phase 3 Multicenter Study of Revusiran in Patients with Hereditary Transthyretin-Mediated (hATTR) Amyloidosis with Cardiomyopathy (ENDEAVOUR). Cardiovasc Drugs Ther (2020). paper; opened 2026-11-17 all records citing this
History of this record (5)

This record has been changed 5 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.

Order
5
Stage
hit-discovery
Status
partly written
Last updated
2026-11-17
Record
content/cascade/hit-discovery.json