Molecular Representations in Machine-Learning-Based Prediction of PK Parameters for Insulin Analogs.

Therapeutic peptides and proteins derived from either endogenous hormones, such as insulin, or de novo design via display technologies occupy a distinct pharmaceutical space in between small molecules and large proteins such as antibodies. Optimizing the pharmacokinetic (PK) profile of drug candidates is of high importance when it comes to prioritizing lead candidates, and machine-learning models can provide a relevant tool to accelerate the drug design process. Predicting PK parameters of proteins remains difficult due to the complex factors that influence PK properties; furthermore, the data sets are small compared to the variety of compounds in the protein space.

Laboratory performance prediction using virtual reality behaviometrics.

In this study, we show that virtual reality (VR) behaviometrics can be used for the assessment of compliance and physical laboratory skills. Drawing on approaches from machine learning and classical statistics, significant behavioral predictors were deduced from a logistic regression model that classified students and biopharma company employees as experts or novices on pH meter handling with 77% accuracy. Specifically, the game score and number of interactions in VR tasks requiring practical skills were found to be performance predictors.