Potential of Artificial Intelligence to Accelerate Drug Development for Rare Diseases.

The growth in breadth and depth of artificial intelligence (AI) applications has been fast, running hand in hand with the increasing amount of digital data available. Here, we comment on the application of AI in the field of drug development, with a strong focus on the specific achievements and challenges posed by rare diseases. Data paucity and high costs make drug development for rare diseases especially hard. AI can enable otherwise inaccessible approaches based on the large-scale integration of heterogeneous datasets and knowledge bases, guided by expert biological understanding. Obstacles still exist for the routine use of AI in the usually conservative pharmaceutical domain, which can easily become disillusioned. It is crucial to acknowledge that AI is a powerful, supportive tool that can assist but not replace human expertise in the various phases and aspects of drug discovery and development.

Druggability Assessment in TRAPP Using Machine Learning Approaches.

Accurate protein druggability predictions are important for the selection of drug targets in the early stages of drug discovery. Because of the flexible nature of proteins, the druggability of a binding pocket may vary due to conformational changes. We have therefore developed two statistical models, a logistic regression model (TRAPP-LR) and a convolutional neural network model (TRAPP-CNN), for predicting druggability and how it varies with changes in the spatial and physicochemical properties of a binding pocket. These models are integrated into TRAnsient Pockets in Proteins (TRAPP), a tool for the analysis of binding pocket variations along a protein motion trajectory. The models, which were trained on publicly available and self-augmented datasets, show equivalent or superior performance to existing methods on test sets of protein crystal structures and have sufficient sensitivity to identify potentially druggable protein conformations in trajectories from molecular dynamics simulations. Visualization of the evidence for the decisions of the models in TRAPP facilitates identification of the factors affecting the druggability of protein binding pockets.

A genetic screen pinpoints ribonucleotide reductase residues that sustain dNTP homeostasis and specifies a highly mutagenic type of dNTP imbalance.

The balance and the overall concentration of intracellular deoxyribonucleoside triphosphates (dNTPs) are important determinants of faithful DNA replication. Despite the established fact that changes in dNTP pools negatively influence DNA replication fidelity, it is not clear why certain dNTP pool alterations are more mutagenic than others. As intracellular dNTP pools are mainly controlled by ribonucleotide reductase (RNR), and given the limited number of eukaryotic RNR mutations characterized so far, we screened for RNR1 mutations causing mutator phenotypes in Saccharomyces cerevisiae. We identified 24 rnr1 mutant alleles resulting in diverse mutator phenotypes linked in most cases to imbalanced dNTPs. Among the identified rnr1 alleles the strongest mutators presented a dNTP imbalance in which three out of the four dNTPs were elevated (dCTP, dTTP and dGTP), particularly if dGTP levels were highly increased. These rnr1 alleles caused growth defects/lethality in DNA replication fidelity-compromised backgrounds, and caused strong mutator phenotypes even in the presence of functional DNA polymerases and mismatch repair. In summary, this study pinpoints key residues that contribute to allosteric regulation of RNR's overall activity or substrate specificity. We propose a model that distinguishes between different dNTP pool alterations and provides a mechanistic explanation why certain dNTP imbalances are particularly detrimental.

The In Situ Tryptophan Analogue Probes the Conformational Dynamics in Asparaginase Isozymes.

Dynamic water solvation is crucial to protein conformational reorganization and hence to protein structure and functionality. We report here the characterization of water dynamics on the L-asparaginase structural homology isozymes L-asparaginases I (AnsA) and II (AnsB), which are shown via fluorescence spectroscopy and dynamics in combination with molecular dynamics simulation to have distinct catalytic activity. By use of the tryptophan (Trp) analog probe 2,7-diaza-tryptophan ((2,7-aza)Trp), which exhibits unique water-catalyzed proton-transfer properties, AnsA and AnsB are shown to have drastically different local water environments surrounding the single Trp. In AnsA, (2,7-aza)Trp exhibits prominent green N(7)-H emission resulting from water-catalyzed excited-state proton transfer. In stark contrast, the N(7)-H emission is virtually absent in AnsB, which supports a water-accessible and a water-scant environment in the proximity of Trp for AnsA and AnsB, respectively. In addition, careful analysis of the emission spectra and corresponding relaxation dynamics, together with the results of molecular dynamics simulations, led us to propose two structural states associated with the rearrangement of the hydrogen-bond network in the vicinity of Trp for the two Ans. The water molecules revealed in the proximity of the Trp residue have semiquantitative correlation with the observed emission spectral variations of (2,7-aza)Trp between AnsA and AnsB. Titration of aspartate, a competitive inhibitor of Ans, revealed an increase in N(7)-H emission intensity in AnsA but no obvious spectral changes in AnsB. The changes in the emission profiles reflect the modulation of structural states by locally confined environment and trapped-water collective motions.