AMPL: A Data-Driven Modeling Pipeline for Drug Discovery

Minnich, Amanda; McLoughlin, Kevin; Tse, Margaret Chui Ling; Deng, Jason; Weber, Andrew; Murad, Neha; Madej, Benjamin D.; Ramsundar, Bharath; Rush, Tom; Calad-Thomson, Stacie; Brase, Jim; Allen, Jonathan

doi:10.48550/arxiv.1911.05211

Cited by 1 publication

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural network and random forest models were trained and evaluated using a data-driven modeling pipeline, AMPL, developed by our group at the ATOM Consortium. 21 Neural networks consisted of one, two or three fully connected hidden layers, with varying numbers of rectified linear unit (ReLU) nodes per layer. During training and evaluation, 30% of nodes were dropped out randomly to avoid overfitting.…”

Section: Model Trainingmentioning

confidence: 99%

“…Our research serves as an application case study for the open source ATOM Modeling PipeLine (AMPL), 21 developed by our team at the Accelerating Therapeutics for Opportunities in Medicine (ATOM) Consortium. In order to facilitate reproduction of our research, we have released example models for BSEP inhibition based on open data and descriptor calculations, which are available on the AMPL GitHub repository (https://github.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Models to Predict Inhibition of the Bile Salt Export Pump

McLoughlin¹,

Jeong²,

Sweitzer³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Drug-induced liver injury (DILI) is the most common cause of acute liver failure and a frequent reason for withdrawal of candidate drugs during preclinical and clinical testing. An important type of DILI is cholestatic liver injury, caused by buildup of bile salts within hepatocytes; it is frequently associated with inhibition of bile salt transporters, such as the bile salt export pump (BSEP). Reliable in silico models to predict BSEP inhibition directly from chemical structures would significantly reduce costs during drug discovery and could help avoid injury to patients. Unfortunately, models published to date have been insufficiently accurate to encourage wide adoption. We report our development of classification and regression models for BSEP inhibition with substantially improved performance over previously published models. Our model development leveraged the ATOM Modeling PipeLine (AMPL) developed by the ATOM Consortium, which enabled us to train and evaluate thousands of candidate models. In the course of model development, we assessed a variety of schemes for chemical featurization, dataset partitioning and class labeling, and identified those producing models that generalized best to novel chemical entities. Our best performing classification model was a neural network with ROC AUC = 0.88 on our internal test dataset and 0.89 on an indepen-dent external compound set. Our best regression model, the first ever reported for predicting BSEP IC50s, yielded a test set R 2 = 0.56 and mean absolute error 0.37, corresponding to a mean 2.3-fold error in predicted IC50s, comparable to experimental variation. These models will thus be useful as inputs to mechanistic predictions of DILI and as part of computational pipelines for drug discovery.

show abstract

Section: Model Trainingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%