Drug discovery and development pipelines are long, complex and depend on numerous factors. Machine learning (ML) approaches provide a set of tools that can improve discovery and decision making for well-specified questions with abundant, high-quality data. Opportunities to apply ML occur in all stages of drug discovery. Examples include target validation, identification of prognostic biomarkers and analysis of digital pathology data in clinical trials. Applications have ranged in context and methodology, with some approaches yielding accurate predictions and insights. The challenges of applying ML lie primarily with the lack of interpretability and repeatability of ML-generated results, which may limit their application. In all areas, systematic and comprehensive high-dimensional data still need to be generated. With ongoing efforts to tackle these issues, as well as increasing awareness of the factors needed to validate ML approaches, the application of ML can promote data-driven decision making and has the potential to speed up the process and reduce failure rates in drug discovery and development.

Dysregulation of Toll-like receptor (TLR) responses to pathogens can lead to pathological inflammation or to immune hyporesponsiveness and susceptibility to infections, and may affect adaptive immune responses. TLRs are therefore attractive therapeutic targets. We assessed the potential of the TLR co-receptor CD14 as a target for therapeutics by investigating the magnitude of its influence on TLR responses. We studied the interaction of CD14 with TLR2 by conducting peptide screening and site-directed mutagenesis analysis and found TLR2 leucine-rich repeats 5, 9, 15, and 20 involved in interaction with CD14. Peptides representing these regions interacted with CD14 and enhanced TLR2- and TLR4-mediated proinflammatory responses to bacterial pathogens in vitro. Notably, the peptides' immune boosting capacity helped to rescue proinflammatory responses of immunosuppressed sepsis patients ex vivo. In vivo, peptide treatment increased phagocyte recruitment and accelerated bacterial clearance in murine models of Gram-negative and Gram-positive bacterial peritonitis. Up-modulating CD14's co-receptor activity with TLR2-derived peptides also enhanced antigen-induced dendritic cell (DC) maturation and interleukin-2 production and, most notably, differentially affected DC cytokine profile upon antigen stimulation, promoting a T helper 1-skewed adaptive immune response. Biochemical, cell imaging, and molecular docking studies showed that peptide binding to CD14 accelerates microbial ligand transfer from CD14 to TLR2, resulting in increased and sustained ligand occupancy of TLR2 and receptor clustering for signaling. These findings reveal the influence that CD14 exerts on TLR activities and describe a potential therapeutic strategy to amplify responses to different pathogens mediated by different TLRs by targeting the common TLR co-receptor, CD14.

We have recently described a method based on artificial neural networks to cluster protein sequences into families. The network was trained with Kohonen's unsupervised learning algorithm using, as inputs, the matrix patterns derived from the dipeptide composition of the proteins. We present here a large-scale application of that method to classify the 1,758 human protein sequences stored in the SwissProt database (release 19.0), whose lengths are greater than 50 amino acids. In the final 2-dimensional topologically ordered map of 15 X 15 neurons, proteins belonging to known families were associated with the same neuron or with neighboring ones. Also, as an attempt to reduce the time-consuming learning procedure, we compared 2 learning protocols: one of 500 epochs (100 SUN CPU-hours [CPU-h]), and another one of 30 epochs (6.7 CPU-h). A further reduction of learningcomputing time, by a factor of about 3.3, with similar protein clustering results, was achieved using a matrix of 11 x 11 components to represent the sequences. Although network training is time consuming, the classification of a new protein in the final ordered map is very fast (14.6 CPU-seconds). We also show a comparison between the artificial neural network approach and conventional methods of biosequence analysis.

Single-cell technologies, particularly single-cell RNA sequencing (scRNA-seq) methods, together with associated computational tools and the growing availability of public data resources, are transforming drug discovery and development. New opportunities are emerging in target identification owing to improved disease understanding through cell subtyping, and highly multiplexed functional genomics screens incorporating scRNA-seq are enhancing target credentialling and prioritization. ScRNA-seq is also aiding the selection of relevant preclinical disease models and providing new insights into drug mechanisms of action. In clinical development, scRNA-seq can inform decision-making via improved biomarker identification for patient stratification and more precise monitoring of drug response and disease progression. Here, we illustrate how scRNA-seq methods are being applied in key steps in drug discovery and development, and discuss ongoing challenges for their implementation in the pharmaceutical industry.

A new method based on neural networks to cluster proteins into families is described. The network is trained with the Kohonen unsupervised learning algorithm, using matrix pattern representations of the protein sequences as inputs. The components (x, y) of these 20 x 20 matrix patterns are the normalized frequencies of all pairs xy of amino acids in each sequence. We investigate the influence of different learning parameters in the final topological maps obtained with a learning set of ten proteins belonging to three established families. In all cases, except in those where the synaptic vectors remains nearly unchanged during learning, the ten proteins are correctly classified into the expected families. The classification by the trained network of mutated or incomplete sequences of the learned proteins is also analysed. The neural network gives a correct classification for a sequence mutated in 21.5% +/- 7% of its amino acids and for fragments representing 7.5% +/- 3% of the original sequence. Similar results were obtained with a learning set of 32 proteins belonging to 15 families. These results show that a neural network can be trained following the Kohonen algorithm to obtain topological maps of protein sequences, where related proteins are finally associated to the same winner neuron or to neighboring ones, and that the trained network can be applied to rapidly classify new sequences. This approach opens new possibilities to find rapid and efficient algorithms to organize and search for homologies in the whole protein database.