What is Hit Identification and when do you need it?
Hit identification is the point in discovery where you stop theorizing about your target and start finding actual molecules that touch it. You take a validated target and a screening-ready assay, run a library of compounds (or fragments, or DNA-encoded molecules, or a virtual deck) against it, and pull out the actives that bind or modulate the target. The output is not a drug. It is a confirmed, ranked set of hits with clean dose-response data, good enough to hand to a chemistry team without them quietly wondering whether half the list is noise.
You reach this step right after assay development and screening, and just before hit-to-lead. The order matters. If your assay is not reliable (poor Z-prime, drifting controls, a readout that fires on detergents and aggregators), screening it only produces a longer list of artifacts, so most experienced teams will not start a campaign until the assay has been qualified. Hit ID is also where the screening strategy gets chosen: high-throughput screening (HTS) of a large diversity library, fragment-based screening when you want better starting points for a hard target, DNA-encoded library (DEL) selection when you need to interrogate billions of compounds cheaply, or virtual and structure-based screening when you have a crystal structure and want to triage in silico first.
The decision a buyer actually makes here is which screening modality fits the target and what counts as a confirmed hit. A kinase with a known pocket and a co-crystal structure is a very different campaign from an orphan GPCR or a protein-protein interaction with no obvious druggable site. Cell-based phenotypic screening, biochemical screening, and biophysical confirmation (SPR, thermal shift, NMR) each answer a different question, and the right CRO is the one that has run your kind of target before, not just the one with the biggest compound deck.
What does a Hit Identification CRO actually do?
A hit-ID CRO runs the screen and, just as importantly, separates the real hits from the things that only look like hits. The first half is execution: library handling and compound logistics, the primary screen at single concentration or in dose, plate-level quality control, and hit calling against a defined cutoff. The second half is triage, which is where the value sits. Singletons get retested, dose-response curves (IC50 or EC50) get generated, and known frequent hitters, PAINS, aggregators, and assay-interference compounds get flagged and removed before anyone celebrates a number.
Confirmation is the part buyers underestimate. A primary hit is a hypothesis; a confirmed hit has survived orthogonal validation. That usually means an independent assay format to rule out readout-specific artifacts, a counter-screen against a related target or a parallel pathway, and a biophysical binding check (surface plasmon resonance, thermal shift, or fragment NMR) to show the compound actually engages the protein. For fragment and DEL campaigns the CRO also resynthesizes off-DNA hits and confirms activity, since DEL actives are guilty until proven innocent. The deliverable a good vendor hands over is a hit list with structures, confirmed potencies, cluster or chemotype analysis, liability flags, and a clear recommendation of which series are worth taking into hit-to-lead.
How to choose a Hit Identification CRO?
The campaign succeeds or fails on fit to your specific target and a clean handoff into chemistry, not on the size of the logo or the library. Score two or three candidates against the same written scope (the target, the assay, the screening modality, and what you will accept as a confirmed hit) so you are comparing the same work rather than three differently shaped quotes. Below are the checks that actually predict a usable hit list.
- Quality and GxP status: hit ID is research-grade, non-GLP work, so do not pay GLP premiums for it. What you do want is reproducibility and traceability: a qualified assay with documented Z-prime, controls and acceptance criteria, sound electronic-notebook practice, and clear chain of custody on compounds and data.
- Capacity and lead time: ask about the screening slot, current queue, and realistic turnaround from assay-ready to confirmed hits, not just bench time. A great lab booked solid for months can be slower than a good lab with an open deck. Confirm whether dose-response and orthogonal confirmation are inside the quoted timeline or a separate follow-on.
- Modality and indication fit: match the screening approach to the target. HTS for tractable targets, fragment-based or DEL for hard ones, virtual or structure-based screening when you have a structure. Ask for case studies on your target class (kinase, GPCR, protease, PPI, ion channel) and confirm the scientists who would run it have done that exact kind of campaign.
- Region and regulatory track record: hit ID does not enter a regulatory filing, so location is mainly about cost, time zone, and IP jurisdiction rather than FDA inspection history. Confirm the jurisdiction where compounds and data are handled and that it aligns with where you intend to file and protect IP later.
- Data quality: insist on what a confirmed hit means up front. You want orthogonal confirmation, counter-screens against frequent hitters and aggregators, biophysical binding evidence, full dose-response curves, and the raw data, not a one-page summary of survivors. Honest reporting of what failed is as important as the hits that passed.
- IP and confidentiality: settle ownership before any work starts. The buyer should own the hits and the screening results from a funded campaign. Watch for vendors with proprietary libraries (DEL or virtual platforms) who may claim rights to platform-derived hits, and get compound transfer, data delivery, and confidentiality on the target in writing.