Experimental design and statistical analysis in high throughput screening
Bibliographic record
Abstract
High throughput screening (HTS) is a biotechnology that allows researchers to detect the small number of active features (e.g. small molecules, small interfering RNAs) among libraries containing up to hundreds of thousands of features. HTS assays, as with all experimental techniques, are prone to both random error resulting from the inherent variability of biological processes or experimental procedures, and systematic error which can be introduced through any number of known or unknown sources. The effect of both types of error can result in truly inactive features being labeled as active (false positives) and truly active features being labeled as inactive (false negatives). The goal of experimental design and statistical analysis is to minimize and estimate the error of an assay, although in the HTS field these methods are not always fully utilized.This thesis presents improvements in the statistical analysis and experimental design of HTS in order to improve the detection of rare biological activity. I first present a comparison of the effectiveness of normalization methods for HTS screening in two titration series experiments and extend the results in a third experiment with two differently designed but otherwise identical screens: compounds in replicate plates were either placed in the same well locations or were randomly assigned to different locations. Best results were obtained with a combination of appropriate normalization and randomization. Secondly, the Single Assay-wide Variance Experimental (SAVE) design is introduced whereby a small replicated subset of an entire screen is used to derive Empirical Bayes random error estimates which are applied to the remaining majority of unreplicated measurements. SAVE is shown to produce valid and informative P-values comparable to the P-values produced with multi-replicate data. Thirdly, the Control Plate Regression (CPR) normalization method, designed for assays such as secondary screens where there may be a majority of active features, is developed and shown to outperform current methodology. Diagnostic techniques are provided that allow researchers to predict the effectiveness and appropriateness of applying CPR. Lastly, the Statistics and dIagnostic Graphs for HTS (SIGHTS) software was developed to implement many of the techniques discussed in this thesis and is designed to be accessible to researchers with no programming experience.Combining graphical assessments, randomization procedures, normalization methods customized to the requirements of the screen, and statistical testing is shown to produce superior results to current HTS analysis techniques.
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 | 0.120 | 0.293 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.004 | 0.004 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.002 | 0.005 |
| Scholarly communication | 0.004 | 0.002 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.004 | 0.008 |
| Insufficient payload (model declined to judge) | 0.011 | 0.003 |
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".