High-throughput crystallography for rapid fragment growth from crude arrays by low-cost robotics
Bibliographic record
Abstract
We demonstrate that a simple workflow of array synthesis, combining low-cost robotics with analytic techniques to deconvolute crude reaction mixtures, is an effective way to collect structural data on a binding site. Starting from the high information content of the crystallographic fragment screens on PHIP(2) (second bromodomain of the pleckstrin homology domain interacting protein), a collection of more than 1800 compounds was enumerated. Several thousand Crude Reaction Mixtures (CRMs) were synthesized on one robotic platform, an OpenTrons OT-1 liquid handler, using reaction sequences of up to 5 chemical steps. Analysis via MScheck, an algorithm-based system for finding an m/z in a CRM, significantly shortened product identification protocol times. 969 usable X-ray diffraction datasets were acquired, which resolved as 22 reaction products binding to the protein, 19 with conserved poses relative to the original fragment and 3 with a new, unexpected binding pose. The 22 crystallographic hit compounds were subsequently tested with peptide displacement alpha-screen assay and time-resolved grating-coupled interferometry-based biosensor assays, which confirmed one molecule with an IC50 = 34 μM and KD = 50 μM, from an inactive fragment. The procedures described are entirely formulaic and engineerable and the method is eminently scalable. We anticipate that this cheap, low solvent-use approach will yield vast amounts of data, enabling rapid structural SAR landscape exploration around fragments, leading to faster fragment-to-lead times.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".