Predicting the impact of sequence motifs on gene regulation using single-cell data.

The binding of transcription factors at proximal promoters and distal enhancers is central to gene regulation. Identifying regulatory motifs and quantifying their impact on expression remains challenging. Using a convolutional neural network trained on single-cell data, we infer putative regulatory motifs and cell type-specific importance. Our model, scover, explains 29% of the variance in gene expression in multiple mouse tissues. Applying scover to distal enhancers identified using scATAC-seq from the developing human brain, we identify cell type-specific motif activities in distal enhancers. Scover can identify regulatory motifs and their importance from single-cell data where all parameters and outputs are easily interpretable.

EvoAug: improving generalization and interpretability of genomic deep neural networks with evolution-inspired data augmentations.

Deep neural networks (DNNs) hold promise for functional genomics prediction, but their generalization capability may be limited by the amount of available data. To address this, we propose EvoAug, a suite of evolution-inspired augmentations that enhance the training of genomic DNNs by increasing genetic variation. Random transformation of DNA sequences can potentially alter their function in unknown ways, so we employ a fine-tuning procedure using the original non-transformed data to preserve functional integrity. Our results demonstrate that EvoAug substantially improves the generalization and interpretability of established DNNs across prominent regulatory genomics prediction tasks, offering a robust solution for genomic DNNs.

Predicting protein-ligand interactions based on bow-pharmacological space and Bayesian additive regression trees.

Identifying potential protein-ligand interactions is central to the field of drug discovery as it facilitates the identification of potential novel drug leads, contributes to advancement from hits to leads, predicts potential off-target explanations for side effects of approved drugs or candidates, as well as de-orphans phenotypic hits. For the rapid identification of protein-ligand interactions, we here present a novel chemogenomics algorithm for the prediction of protein-ligand interactions using a new machine learning approach and novel class of descriptor. The algorithm applies Bayesian Additive Regression Trees (BART) on a newly proposed proteochemical space, termed the bow-pharmacological space. The space spans three distinctive sub-spaces that cover the protein space, the ligand space, and the interaction space. Thereby, the model extends the scope of classical target prediction or chemogenomic modelling that relies on one or two of these subspaces. Our model demonstrated excellent prediction power, reaching accuracies of up to 94.5-98.4% when evaluated on four human target datasets constituting enzymes, nuclear receptors, ion channels, and G-protein-coupled receptors . BART provided a reliable probabilistic description of the likelihood of interaction between proteins and ligands, which can be used in the prioritization of assays to be performed in both discovery and vigilance phases of small molecule development.

Prebiotic selection for motifs in a model of template-free elongation of polymers within compartments.

The transition from prelife where self-replication does not occur, to life which exhibits self-replication and evolution, has been a subject of interest for many decades. Membranes, forming compartments, seem to be a critical component of this transition as they provide several concurrent benefits. They maintain localized interactions, generate electro-chemical gradients, and help in selecting cooperative functions as they arise. These functions pave the way for the emergence and maintenance of simple metabolic cycles and polymers. In the context of origin of life, evolution of information-carrying molecules and RNA based enzymes within compartments has been subject to intensive theoretical and experimental research. Hence, many experimental efforts aim to produce compartments that contain elongating polynucleotides (also referred to as protocells), which store information and perform catalysis. Despite impressive experimental progress, we are still relatively ignorant about the dynamics by which elongating polynucleotides can produce more sophisticated behaviors. Here we perform computer simulations to couple information production through template-free elongation of polymers with dividing compartments. We find that polymers with a simple ability-biasing the concentration of monomers within their own compartment-can acquire a selective advantage in prelife. We further investigate whether such a mechanism allows for cooperative dynamics to dominate over purely competitive ones. We show that under this system of biased monomer addition, even without template-directed self-replication, genetic motifs can emerge, compete, cooperate, and ultimately survive within the population.