In silico ADME/tox comes of age: twenty years later

Near-Term Quantum Classification Algorithms Applied to Antimalarial Drug Discovery

Computational approaches are widely applied in drug discovery to explore properties related to bioactivity, physiochemistry, and toxicology. Over at least the last 20 years, the exploitation of machine learning on molecular data sets has been used to understand the structure-activity relationships that exist between biomolecules and druggable targets. More recently, these methods have also seen application for phenotypic screening data for neglected diseases such as tuberculosis and malaria. Herein, we apply machine learning to build quantum Quantitative Structure Activity Relationship models from antimalarial data sets. There is a continual need for new antimalarials to address drug resistance, and the readily available in vitro data sets could be utilized with newer machine learning approaches as these develop. Furthermore, quantum machine learning is a relatively new method that uses a quantum computer to perform the calculations. First, we present a classical-quantum hybrid computational approach by building a Latent Bernoulli Autoencoder machine learning model for compressing bit-vector descriptors to a size that can be adapted to quantum computers for classification tasks with limited loss of embedded information. Second, we apply our method for feature map compression to quantum classification algorithms, including a completely novel machine learning algorithm with no analogy in classical computers: the Quantum Fourier Transform Classifier. We apply both these approaches to build quantum machine learning models for small-molecule antimalarials with quantum simulation software and then benchmark these quantum models against classical machine learning approaches. While there are many challenges currently facing the development of reliable quantum computers, our results demonstrate that there is potential for the use of this technology in the field of drug discovery.

Application of machine learning models for property prediction to targeted protein degraders

Machine learning (ML) systems can model quantitative structure-property relationships (QSPR) using existing experimental data and make property predictions for new molecules. With the advent of modalities such as targeted protein degraders (TPD), the applicability of QSPR models is questioned and ML usage in TPD-centric projects remains limited. Herein, ML models are developed and evaluated for TPDs' property predictions, including passive permeability, metabolic clearance, cytochrome P450 inhibition, plasma protein binding, and lipophilicity. Interestingly, performance on TPDs is comparable to that of other modalities. Predictions for glues and heterobifunctionals often yield lower and higher errors, respectively. For permeability, CYP3A4 inhibition, and human and rat microsomal clearance, misclassification errors into high and low risk categories are lower than 4% for glues and 15% for heterobifunctionals. For all modalities, misclassification errors range from 0.8% to 8.1%. Investigated transfer learning strategies improve predictions for heterobifunctionals. This is the first comprehensive evaluation of ML for the prediction of absorption, distribution, metabolism, and excretion (ADME) and physicochemical properties of TPD molecules, including heterobifunctional and molecular glue sub-modalities. Taken together, our investigations show that ML-based QSPR models are applicable to TPDs and support ML usage for TPDs' design, to potentially accelerate drug discovery.

The Goldilocks paradigm: comparing classical machine learning, large language models, and few-shot learning for drug discovery applications

Recent advances in machine learning (ML) have led to newer model architectures including transformers (large language models, LLMs) showing state of the art results in text generation and image analysis as well as few-shot learning (FSLC) models which offer predictive power with extremely small datasets. These new architectures may offer promise, yet the 'no-free lunch' theorem suggests that no single model algorithm can outperform at all possible tasks. Here, we explore the capabilities of classical (SVR), FSLC, and transformer models (MolBART) over a range of dataset tasks and show a 'goldilocks zone' for each model type, in which dataset size and feature distribution (i.e. dataset "diversity") determines the optimal algorithm strategy. When datasets are small ( < 50 molecules), FSLC tend to outperform both classical ML and transformers. When datasets are small-to-medium sized (50-240 molecules) and diverse, transformers outperform both classical models and few-shot learning. Finally, when datasets are of larger and of sufficient size, classical models then perform the best, suggesting that the optimal model to choose likely depends on the dataset available, its size and diversity. These findings may help to answer the perennial question of which ML algorithm is to be used when faced with a new dataset.

Drug Selection and Posology, Optimal Therapies and Risk/Benefit Assessment in Medicine: The Paradigm of Iron-Chelating Drugs

The design of clinical protocols and the selection of drugs with appropriate posology are critical parameters for therapeutic outcomes. Optimal therapeutic protocols could ideally be designed in all diseases including for millions of patients affected by excess iron deposition (EID) toxicity based on personalised medicine parameters, as well as many variations and limitations. EID is an adverse prognostic factor for all diseases and especially for millions of chronically red-blood-cell-transfused patients. Differences in iron chelation therapy posology cause disappointing results in neurodegenerative diseases at low doses, but lifesaving outcomes in thalassemia major (TM) when using higher doses. In particular, the transformation of TM from a fatal to a chronic disease has been achieved using effective doses of oral deferiprone (L1), which improved compliance and cleared excess toxic iron from the heart associated with increased mortality in TM. Furthermore, effective L1 and L1/deferoxamine combination posology resulted in the complete elimination of EID and the maintenance of normal iron store levels in TM. The selection of effective chelation protocols has been monitored by MRI T2* diagnosis for EID levels in different organs. Millions of other iron-loaded patients with sickle cell anemia, myelodysplasia and haemopoietic stem cell transplantation, or non-iron-loaded categories with EID in different organs could also benefit from such chelation therapy advances. Drawbacks of chelation therapy include drug toxicity in some patients and also the wide use of suboptimal chelation protocols, resulting in ineffective therapies. Drug metabolic effects, and interactions with other metals, drugs and dietary molecules also affected iron chelation therapy. Drug selection and the identification of effective or optimal dose protocols are essential for positive therapeutic outcomes in the use of chelating drugs in TM and other iron-loaded and non-iron-loaded conditions, as well as general iron toxicity.