Dataset Information

Changing the HTS Paradigm: AI-Driven Iterative Screening for Hit Finding.

ABSTRACT: Iterative screening is a process in which screening is done in batches, with each batch filled by using machine learning to select the most promising compounds from the library based on the previous results. We believe iterative screening is poised to enhance the screening process by improving hit finding while at the same time reducing the number of compounds screened. In addition, we see this process as a key enabler of next-generation high-throughput screening (HTS), which uses more complex assays that better describe the biology but demand more resource per screened compound. To demonstrate the utility of these methods, we retrospectively analyze HTS data from PubChem with a focus on machine learning-based screening strategies that can be readily implemented in practice. Our results show that over a variety of HTS experimental paradigms, an iterative screening setup that screens a total of 35% of the screening collection over as few as three iterations has a median return rate of approximately 70% of the active compounds. Increasing the portion of the library screened to 50% yields median returns of approximately 80% of actives. Using six iterations increases these return rates to 78% and 90%, respectively. The best results were achieved with machine learning models that can be run on a standard desktop. By demonstrating that the utility of iterative screening holds true even with a small number of iterations, and without requiring significant computational resources, we provide a roadmap for the practical implementation of these techniques in hit finding.

SUBMITTER: Dreiman GHS

PROVIDER: S-EPMC7838329 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Changing the HTS Paradigm: AI-Driven Iterative Screening for Hit Finding.

Dreiman Gabriel H S GHS Bictash Magda M Fish Paul V PV Griffin Lewis L Svensson Fredrik F

SLAS discovery : advancing life sciences R & D 20200818 2

Iterative screening is a process in which screening is done in batches, with each batch filled by using machine learning to select the most promising compounds from the library based on the previous results. We believe iterative screening is poised to enhance the screening process by improving hit finding while at the same time reducing the number of compounds screened. In addition, we see this process as a key enabler of next-generation high-throughput screening (HTS), which uses more complex a ...[more]

PMID: 32808550

Similar Datasets

Project description:ObjectivesDiagnostic accuracy was evaluated for various screening tools, including mobile health (mHealth) pure-tone screening, tympanometry, distortion product otoacoustic emissions (DPOAE), and inclusion of high frequencies to determine the most accurate screening protocol for identifying children with hearing loss in rural Alaska where the prevalence of middle ear disease is high.DesignHearing screening data were collected as part of two cluster randomized trials conducted in 15 communities in rural northwest Alaska. All children enrolled in school from preschool to 12th grade were eligible. Analysis was limited to data collected 2018 to 2019 (n = 1449), when both trials were running and measurement of high frequencies were included in the protocols. Analyses included estimates of diagnostic accuracy for each screening tool, as well as exploring performance by age and grade. Multiple imputation was used to assess diagnostic accuracy in younger children, where missing data were more prevalent due to requirements for conditioned responses. The audiometric reference standard included otoscopy, tympanometry, and high frequencies to ensure detection of infection-related and noise-induced hearing loss.ResultsBoth the mHealth pure-tone screen and DPOAE screen performed better when tympanometry was added to the protocol (increase in sensitivity of 19.9%, 95% Confidence Interval (CI): 15.9 to 24.1 for mHealth screen, 17.9%, 95% CI: 14.0 to 21.8 for high-frequency mHealth screen, and 10.4%, 95% CI: 7.5 to 13.9 for DPOAE). The addition of 6 kHz to the mHealth pure-tone screen provided an 8.7 percentage point improvement in sensitivity (95% CI: 6.5 to 11.3). Completeness of data for both the reference standard and the mHealth screening tool differed substantially by age, due to difficulty with behavioral testing in young children. By age 7, children were able to complete behavioral testing, and data indicated that high-frequency mHealth pure-tone screen with tympanometry was the superior tool for children 7 years and older. For children 3 to 6 years of age, DPOAE plus tympanometry performed the best, both for complete data and multiply imputed data, which better approximates accuracy for children with missing data.ConclusionsThis study directly evaluated pure-tone, DPOAE, and tympanometry tools as part of school hearing screening in rural Alaskan children (3 to 18+ years). Results from this study indicate that tympanometry is a key component in the hearing screening protocol, particularly in environments with higher prevalence of infection-related hearing loss. DPOAE is the preferred hearing screening tool when evaluating children younger than 7 years of age (below 2nd grade in the United States) due to the frequency of missing data with behavioral testing in this age group. For children 7 years and older, the addition of high frequencies to pure-tone screening increased the accuracy of screening, likely due to improved identification of hearing loss from noise exposure. The lack of a consistent reference standard in the literature makes comparing across studies challenging. In our study with a reference standard inclusive of otoscopy, tympanometry, and high frequencies, less than ideal sensitivities were found even for the most sensitive screening protocols, suggesting more investigation is necessary to ensure screening programs are appropriately identifying noise- and infection-related hearing loss in rural, low-resource settings.

Project description:Quantitative high throughput screening (qHTS) assays use cells or tissues to screen thousands of compounds in a short period of time. Data generated from qHTS assays are then evaluated using nonlinear regression models, such as the Hill model, and decisions regarding toxicity are made using the estimates of the parameters of the model. For any given compound, the variability in the observed response may either be constant across dose groups (homoscedasticity) or vary with dose (heteroscedasticity). Since thousands of compounds are simultaneously evaluated in a qHTS assay, it is not practically feasible for an investigator to perform residual analysis to determine the variance structure before performing statistical inferences on each compound. Since it is well-known that the variance structure plays an important role in the analysis of linear and nonlinear regression models it is therefore important to have practically useful and easy to interpret methodology which is robust to the variance structure. Furthermore, given the number of chemicals that are investigated in the qHTS assay, outliers and influential observations are not uncommon. In this article we describe preliminary test estimation (PTE) based methodology which is robust to the variance structure as well as any potential outliers and influential observations. Performance of the proposed methodology is evaluated in terms of false discovery rate (FDR) and power using a simulation study mimicking a real qHTS data. Of the two methods currently in use, our simulations studies suggest that one is extremely conservative with very small power in comparison to the proposed PTE based method whereas the other method is very liberal. In contrast, the proposed PTE based methodology achieves a better control of FDR while maintaining good power. The proposed methodology is illustrated using a data set obtained from the National Toxicology Program (NTP). Additional information, simulation results, data and computer code are available online as supplementary materials.

Dataset Information

Changing the HTS Paradigm: AI-Driven Iterative Screening for Hit Finding.

Publications

Changing the HTS Paradigm: AI-Driven Iterative Screening for Hit Finding.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets