What is Assay Development & Screening and when do you need it?
Assay development is where you build the measuring stick for a drug program. Before you can ask whether a compound does anything useful, you need a readout that fires reliably when the target is engaged and stays quiet when it is not. The assay can be biochemical (a purified enzyme or binding interaction read by fluorescence, luminescence, TR-FRET, AlphaScreen, or a mobility shift), cell-based (a reporter gene, a calcium flux, a viability or proliferation readout, or a high-content imaging endpoint), or phenotypic (looking for a disease-relevant change in a cell without naming the target up front). Whichever format, the work is the same in spirit: design it, optimize the reagents and conditions, then qualify that it performs. Teams usually gate on a Z-prime above 0.5 and a tight signal window, with day-to-day and plate-to-plate reproducibility checked before anyone screens a single library plate.
Screening is what the assay was built for. Once the readout is solid, you run it against compound collections to find starting points. That might be high-throughput screening across a diversity library of tens of thousands to a few million compounds, a focused or targeted library of known chemotypes, fragment-based screening read by SPR or thermal shift, or a phenotypic campaign. The output is a hit list, plus the confirmation work that separates real activity from assay artifacts: dose-response to get IC50 or EC50, counter-screens to catch aggregators and frequent hitters, and orthogonal assays to confirm the mechanism is what you think it is.
You reach this stage right after target identification and validation, and just before hit-to-lead. In practical terms, you need it the moment you have a target you believe in but no chemical matter to work on, and no trustworthy way to rank compounds. Skipping the development part and rushing to screen is the classic expensive mistake: a poorly behaved assay produces a hit list full of artifacts, and you pay for that downstream in months of chemistry chasing molecules that were never real.
What does an Assay Development & Screening CRO actually do?
A good assay CRO does more than run plates. The job starts with assay design and reagent work: producing or sourcing the target protein, picking the format that fits the biology and the budget, and miniaturizing from a benchtop format down to 384-well or 1536-well so the screen is affordable at scale. Then comes optimization and qualification, where they tune buffer, substrate, DMSO tolerance, incubation, and read conditions, establish the assay statistics, and run a pilot or validation set before committing the full library.
On the screening side, the CRO handles liquid handling and automation, the primary screen itself, hit picking, and the confirmation cascade. The deliverable you actually care about is not raw plate data but a curated, confirmed hit list with dose-response curves, counter-screen results, and a clear account of what was filtered and why. Many will also support secondary and mechanistic assays, selectivity panels, and biophysical confirmation by SPR, ITC, or thermal shift so you carry forward hits that are genuinely engaging the target.
- Assay design across biochemical, cell-based, reporter, high-content, and phenotypic formats
- Reagent generation and target protein supply, plus assay miniaturization to 384 and 1536-well
- Optimization and qualification (Z-prime, signal window, DMSO tolerance, reproducibility)
- High-throughput, focused-library, and fragment-based screening with automated liquid handling
- Hit confirmation: dose-response (IC50/EC50), counter-screens, orthogonal and selectivity assays
- A curated, confirmed hit list with the data and triage rationale, ready for hit-to-lead
How to choose an Assay Development & Screening CRO?
The first question is fit to your specific target and format, not the size of the screening deck. A CRO that runs flawless biochemical kinase HTS may be the wrong partner for a tricky membrane-protein cell assay or a phenotypic imaging campaign. Ask for relevant case studies in your target class and assay type, and confirm the scientists who would run your program have built that kind of readout before, not just operated the robots. The checklist below covers what separates a clean engagement from a frustrating one.
Most discovery screening is research-grade rather than GLP, so the quality bar here is reproducibility and traceability, qualified assays with documented performance and clean data capture, rather than a regulatory quality system. Confirm the work matches your need before you pay for compliance you do not require, or skip rigor you do.
- Quality and documentation: assay qualification to defined acceptance criteria (Z-prime, reproducibility), good electronic-notebook and data-integrity practice; full GLP is rarely needed at this stage, so do not overpay for it
- Capacity and lead time: current screening queue and realistic timeline for assay development, the primary screen, and confirmation, since a great deck booked solid can be slower than a good one with an open slot
- Modality and target fit: relevant experience with your target class (enzyme, GPCR, ion channel, protein-protein interaction) and the assay format and library type that actually suits the biology
- Region and regulatory track record: where the work is run and whether their standards align with how you will use the data downstream
- Data quality and deliverables: a confirmed, curated hit list with dose-response and counter-screen data, transparent reporting of what was filtered and why, not just a dump of raw plate reads
- IP and confidentiality: clear terms that you own the hits and data arising from the funded screen, plus handling of a target you may not want disclosed and any rights to library or platform-derived compounds