Protein–protein interactions (PPIs) are fundamental processes governing cellular functions, crucial for understanding biological systems at the molecular level. Compared to experimental methods for PPI prediction and site identification, computational deep learning approaches represent an affordable and efficient solution to tackle these problems. Since protein structure can be summarized as a graph, graph neural networks (GNNs) represent the ideal deep learning architecture for the task. In this work, PPI prediction is modeled as a node-focused binary classification task using a GNN to determine whether a generic residue is part of the interface. Biological data were obtained from the Protein Data Bank in Europe (PDBe), leveraging the Protein Interfaces, Surfaces, and Assemblies (PISA) service. To gain a deeper understanding of how proteins interact, the data obtained from PISA were assembled into three datasets: Whole, Interface, and Chain, consisting of data on the whole protein, couples of interacting chains, and single chains, respectively. These three datasets correspond to three different nuances of the problem: identifying interfaces between protein complexes, between chains of the same protein, and interface regions in general. The results indicate that GNNs are capable of solving each of the three tasks with very good performance levels.

Background: The ideal of personalized medicine is to support the clinical decision process towards the right drug for the right patient at the right time, by using, among other diagnostic tools, molecular biomarkers that are specifically dependent on the patient status and on the therapeutic options. Several challenges must be overcome to realize this vision. Patients present a wide spectrum of genetic variability even before developing diseases, and disease like cancer add an extra layer of mutations, while only a very small fraction of such variants have diagnostic or prognostic value. Moreover it is also challenging to predict how the patient will respond to a specific drug based on the patient’s ‘omic’ profiling, since any drug introduces further perturbations in the biochemical model. Methods: In this paper we propose the method Personalized-DrugRank for joint prediction of therapy response and time-to-response for cancer patients undergoing pharmacological therapy after surgery. The method is based on personalizing the DrugMerge methodology for drug repositioning in order to extract a few synthetic indices useful as input to ML prediction tools. In particular the proposed methodology is a novel and principled approach to merging independent patient-specific transcriptomic data with drug perturbation data from cell line assays. One of the key novel features of our approach over the state of the art is the joint prediction of the response of the patient to therapy along with an estimate of the time-to-response (i.e the prediction of the time needed for the therapy to succeed or fail). Findings: We tested our methodology on data from the TCGA (The Cancer Genome Atlas) Program for three cancer types (Breast, Stomach and Colorectal cancer), 10 pharmacological regimens and 13 homogeneous cohorts. For the therapy response prediction task we developed models that attain an average AUC performance 0.749, average pvalue 0.030, average accuracy 0.809 with balanced Positive and Negative Predicting Values. For the time-to-event prediction task we developed regression models for the 13 homogeneous cohorts that attain an average (geometric) Concordance Index performance 0.782 (max 0.904, min 0.651) with average log likelihood pvalue 0.004, improving in nine cohorts over 13 upon models based only on clinical parameters having average Concordance Index 0.678 and average p-value 0.006. Interestingly, we attain statistical significant results even with quite small therapy-homogeneous cohorts (ranging from a minimum of 7 patients to a maximum of 32). Conclusions: The ability of predicting with high accuracy the response of a cancer patient to a chosen pharmacological therapeutic regimen along with an estimate of the time-to-response helps adapting the clinical decision process to the specific patient profile, thus increasing the likelihood of providing correct and timely therapeutic decisions.

Protein–protein interfaces (PPIs) are essential for numerous biological processes and represent attractive targets for drug discovery and protein engineering. Although recent protein language models have enabled powerful sequence representations for downstream structural prediction tasks, they require substantial computational resources and produce general-purpose embeddings that are not optimized for specific applications. In this work, we investigate whether task-specific representation learning can provide a computationally efficient alternative for PPI prediction. Proteins are represented as graphs of secondary structure elements (SSEs), where each node corresponds to an SSE described by its ordered amino acid sequence and edges encode spatial relationships between neighboring SSEs. We propose an approach based on a hybrid Bidirectional Long Short-Term Memory–Graph Neural Network (BiLSTM-GNN) architecture that learns SSE embeddings directly from residue-level physicochemical descriptors and optimizes them jointly with a downstream graph neural network. The proposed approach is compared with an alternative pipeline based on pretrained ESM2 embeddings while keeping the graph representation and GNN architecture unchanged. Experimental results show that the task-specific BiLSTM encoder achieves competitive predictive performance with respect to the pretrained protein language model, in almost all the considered metrics, in the most challenging evaluation setting while requiring a lower computational effort. These findings suggest that end-to-end task-specific representation learning constitutes an effective and efficient alternative to pretrained protein language models for PPI prediction.