Isidro Cortés‐Ciriano

Isidro Cortés-Ciriano, Andreas Bender

Estimating the reliability of individual predictions is key to increase the adoption of computational models and artificial intelligence in preclinical drug discovery, as well as to foster its application to guide decision making in clinical settings. Among the large number of algorithms developed over the last decades to compute prediction errors, Conformal Prediction (CP) has gained increasing attention in the computational drug discovery community. A major reason for its recent popularity is the ease of interpretation of the computed prediction errors in both classification and regression tasks. For instance, at a confidence level of 90% the true value will be within the predicted confidence intervals in at least 90% of the cases. This so called validity of conformal predictors is guaranteed by the robust mathematical foundation underlying CP. The versatility of CP relies on its minimal computational footprint, as it can be easily coupled to any machine learning algorithm at little computational cost. In this review, we summarize underlying concepts and practical applications of CP with a particular focus on virtual screening and activity modelling, and list open source implementations of relevant software. Finally, we describe the current limitations in the field, and provide a perspective on future opportunities for CP in preclinical and clinical drug discovery.

Oliver P Watson, Isidro Cortes-Ciriano, James A Watson

Supervised learning models, also known as quantitative structure-activity regression (QSAR) models, are increasingly used in assisting the process of preclinical, small molecule drug discovery. The models are trained on data consisting of a finite dimensional representation of molecular structures and their corresponding target specific activities. These models can then be used to predict the activity of previously unmeasured novel compounds. In this work we address two problems related to this approach. The first is to estimate the extent to which the quality of the model predictions degrades for compounds very different from the compounds in the training data. The second is to adjust for the screening dependent selection bias inherent in many training data sets. In the most extreme cases, only compounds which pass an activity-dependent screening are reported. By using a semi-supervised learning framework, we show that it is possible to make predictions which take into account the similarity of the testing compounds to those in the training data and adjust for the reporting selection bias. We illustrate this approach using publicly available structure-activity data on a large set of compounds reported by GlaxoSmithKline (the Tres Cantos AntiMalarial Set) to inhibit in vitro P. falciparum growth.

Isidro Cortes-Ciriano, Andreas Bender

While the use of deep learning in drug discovery is gaining increasing attention, the lack of methods to compute reliable errors in prediction for Neural Networks prevents their application to guide decision making in domains where identifying unreliable predictions is essential, e.g. precision medicine. Here, we present a framework to compute reliable errors in prediction for Neural Networks using Test-Time Dropout and Conformal Prediction. Specifically, the algorithm consists of training a single Neural Network using dropout, and then applying it N times to both the validation and test sets, also employing dropout in this step. Therefore, for each instance in the validation and test sets an ensemble of predictions were generated. The residuals and absolute errors in prediction for the validation set were then used to compute prediction errors for test set instances using Conformal Prediction. We show using 24 bioactivity data sets from ChEMBL 23 that dropout Conformal Predictors are valid (i.e., the fraction of instances whose true value lies within the predicted interval strongly correlates with the confidence level) and efficient, as the predicted confidence intervals span a narrower set of values than those computed with Conformal Predictors generated using Random Forest (RF) models. Lastly, we show in retrospective virtual screening experiments that dropout and RF-based Conformal Predictors lead to comparable retrieval rates of active compounds. Overall, we propose a computationally efficient framework (as only N extra forward passes are required in addition to training a single network) to harness Test-Time Dropout and the Conformal Prediction framework, and to thereby generate reliable prediction errors for deep Neural Networks.

Isidro Cortes-Ciriano, Andreas Bender

Deep learning architectures have proved versatile in a number of drug discovery applications, including the modelling of in vitro compound activity. While controlling for prediction confidence is essential to increase the trust, interpretability and usefulness of virtual screening models in drug discovery, techniques to estimate the reliability of the predictions generated with deep learning networks remain largely underexplored. Here, we present Deep Confidence, a framework to compute valid and efficient confidence intervals for individual predictions using the deep learning technique Snapshot Ensembling and conformal prediction. Specifically, Deep Confidence generates an ensemble of deep neural networks by recording the network parameters throughout the local minima visited during the optimization phase of a single neural network. This approach serves to derive a set of base learners (i.e., snapshots) with comparable predictive power on average, that will however generate slightly different predictions for a given instance. The variability across base learners and the validation residuals are in turn harnessed to compute confidence intervals using the conformal prediction framework. Using a set of 24 diverse IC50 data sets from ChEMBL 23, we show that Snapshot Ensembles perform on par with Random Forest (RF) and ensembles of independently trained deep neural networks. In addition, we find that the confidence regions predicted using the Deep Confidence framework span a narrower set of values. Overall, Deep Confidence represents a highly versatile error prediction framework that can be applied to any deep learning-based application at no extra computational cost.