BACKGROUND: Proteins play a pivotal role in the diverse array of biological processes, making the precise prediction of protein-protein interaction (PPI) sites critical to numerous disciplines including biology, medicine and pharmacy. While deep learning methods have progressively been implemented for the prediction of PPI sites within proteins, the task of enhancing their predictive performance remains an arduous challenge.
RESULTS: In this paper, we propose a novel PPI site prediction model (DGCPPISP) based on a dynamic graph convolutional neural network and a two-stage transfer learning strategy. Initially, we implement the transfer learning from dual perspectives, namely feature input and model training that serve to supply efficacious prior knowledge for our model. Subsequently, we construct a network designed for the second stage of training, which is built on the foundation of dynamic graph convolution.
CONCLUSIONS: To evaluate its effectiveness, the performance of the DGCPPISP model is scrutinized using two benchmark datasets. The ensuing results demonstrate that DGCPPISP outshines competing methods in terms of performance. Specifically, DGCPPISP surpasses the second-best method, EGRET, by margins of 5.9%, 10.1%, and 13.3% for F1-measure, AUPRC, and MCC metrics respectively on Dset_186_72_PDB164. Similarly, on Dset_331, it eclipses the performance of the runner-up method, HN-PPISP, by 14.5%, 19.8%, and 29.9% respectively.

Receptor activity-modifying proteins (RAMPs) form complexes with G protein-coupled receptors (GPCRs) and may regulate their cellular trafficking and pharmacology. RAMP interactions have been identified for about 50 GPCRs, but only a few GPCR-RAMP complexes have been studied in detail. To elucidate a comprehensive GPCR-RAMP interactome, we created a library of 215 dual epitope-tagged (DuET) GPCRs representing all GPCR subfamilies and coexpressed each GPCR with each of the three RAMPs. Screening the GPCR-RAMP pairs with customized multiplexed suspension bead array (SBA) immunoassays, we identified 122 GPCRs that showed strong evidence for interaction with at least one RAMP. We screened for interactions in three cell lines and found 23 endogenously expressed GPCRs that formed complexes with RAMPs. Mapping the GPCR-RAMP interactome expands the current system-wide functional characterization of RAMP-interacting GPCRs to inform the design of selective therapeutics targeting GPCR-RAMP complexes.

BACKGROUND: Protein-protein interaction (PPI) networks provide valuable insights into the function of biological systems. Aligning multiple PPI networks may expose relationships beyond those observable by pairwise comparisons. However, assessing the biological quality of multiple network alignments is a challenging problem.
RESULTS: We propose two new measures to evaluate the quality of multiple network alignments using functional information from Gene Ontology (GO) terms. When aligning multiple real PPI networks across species, we observe that both measures are highly correlated with objective quality indicators, such as common orthologs. Additionally, our measures strongly correlate with an alignment\'s ability to predict novel GO annotations, which is a unique advantage over existing GO-based measures.
METHODS: The scripts and the links to the raw and alignment data can be accessed at https://github.com/kimiayazdani/GO_Measures.git.

The African swine fever virus (ASFV) is an often deadly disease in swine and poses a threat to swine livestock and swine producers. With its complex genome containing more than 150 coding regions, developing effective vaccines for this virus remains a challenge due to a lack of basic knowledge about viral protein function and protein-protein interactions between viral proteins and between viral and host proteins. In this work, we identified ASFV-ASFV protein-protein interactions (PPIs) using artificial intelligence-powered protein structure prediction tools. We benchmarked our PPI identification workflow on the Vaccinia virus, a widely studied nucleocytoplasmic large DNA virus, and found that it could identify gold-standard PPIs that have been validated in vitro in a genome-wide computational screening. We applied this workflow to more than 18,000 pairwise combinations of ASFV proteins and were able to identify seventeen novel PPIs, many of which have corroborating experimental or bioinformatic evidence for their protein-protein interactions, further validating their relevance. Two protein-protein interactions, I267L and I8L, I267L__I8L, and B175L and DP79L, B175L__DP79L, are novel PPIs involving viral proteins known to modulate host immune response.

Signaling pathways are responsible for transmitting information between cells and regulating cell growth, differentiation, and death. Proteins in cells form complexes by interacting with each other through specific structural domains, playing a crucial role in various biological functions and cell signaling pathways. Protein-protein interactions (PPIs) within cell signaling pathways are essential for signal transmission and regulation. The spatiotemporal features of PPIs in signaling pathways are crucial for comprehending the regulatory mechanisms of signal transduction. Bimolecular fluorescence complementation (BiFC) is one kind of imaging tool for the direct visualization of PPIs in living cells and has been widely utilized to uncover novel PPIs in various organisms. BiFC demonstrates significant potential for application in various areas of biological research, drug development, disease diagnosis and treatment, and other related fields. This review systematically summarizes and analyzes the technical advancement of BiFC and its utilization in elucidating PPIs within established cell signaling pathways, including TOR, PI3K/Akt, Wnt/β-catenin, NF-κB, and MAPK. Additionally, it explores the application of this technology in revealing PPIs within the plant hormone signaling pathways of ethylene, auxin, Gibberellin, and abscisic acid. Using BiFC in conjunction with CRISPR-Cas9, live-cell imaging, and ultra-high-resolution microscopy will enhance our comprehension of PPIs in cell signaling pathways.

Protein-protein interactions (PPIs) are important for many biological processes, but predicting them from sequence data remains challenging. Existing deep learning models often cannot generalize to proteins not present in the training set and do not provide uncertainty estimates for their predictions. To address these limitations, we present TUnA, a Transformer-based uncertainty-aware model for PPI prediction. TUnA uses ESM-2 embeddings with Transformer encoders and incorporates a Spectral-normalized Neural Gaussian Process. TUnA achieves state-of-the-art performance and, importantly, evaluates uncertainty for unseen sequences. We demonstrate that TUnA\'s uncertainty estimates can effectively identify the most reliable predictions, significantly reducing false positives. This capability is crucial in bridging the gap between computational predictions and experimental validation.

Virus receptors determine the tissue tropism of viruses and have a certain relationship with the clinical outcomes caused by viral infection, which is of great importance for the identification of virus receptors to understand the infection mechanism of viruses and to develop entry inhibitor. Proximity labeling (PL) is a new technique for studying protein-protein interactions, but it has not yet been applied to the identification of virus receptors or co-receptors. Here, we attempt to identify co-receptor of SARS-CoV-2 by employing TurboID-catalyzed PL. The membrane protein angiotensin-converting enzyme 2 (ACE2) was employed as a bait and conjugated to TurboID, and a A549 cell line with stable expression of ACE2-TurboID was constructed. SARS-CoV-2 pseudovirus were incubated with ACE2-TurboID stably expressed cell lines in the presence of biotin and ATP, which could initiate the catalytic activity of TurboID and tag adjacent endogenous proteins with biotin. Subsequently, the biotinylated proteins were harvested and identified by mass spectrometry. We identified a membrane protein, AXL, that has been functionally shown to mediate SARS-CoV-2 entry into host cells. Our data suggest that PL could be used to identify co-receptors for virus entry.

Protein-protein interactions play an important biological role in every aspect of cellular homeostasis and functioning. Proximity labeling mass spectrometry-based proteomics overcomes challenges typically associated with other methods and has quickly become the current state of the art in the field. Nevertheless, tight control of proximity-labeling enzymatic activity and expression levels is crucial to accurately identify protein interactors. Here, we leverage a T2A self-cleaving peptide and a non-cleaving mutant to accommodate the protein of interest in the experimental and control TurboID setup. To allow easy and streamlined plasmid assembly, we built a Golden Gate modular cloning system to generate plasmids for transient expression and stable integration. To highlight our T2A Split/link design, we applied it to identify protein interactions of the glucocorticoid receptor and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) nucleocapsid and non-structural protein 7 (NSP7) proteins by TurboID proximity labeling. Our results demonstrate that our T2A split/link provides an opportune control that builds upon previously established control requirements in the field.

Artificial intelligence has revolutionized the field of protein structure prediction. However, with more powerful and complex software being developed, it is accessibility and ease of use rather than capability that is quickly becoming a limiting factor to end users. LazyAF is a Google Colaboratory-based pipeline which integrates the existing ColabFold BATCH software to streamline the process of medium-scale protein-protein interaction prediction. LazyAF was used to predict the interactome of the 76 proteins encoded on the broad-host-range multi-drug resistance plasmid RK2, demonstrating the ease and accessibility the pipeline provides.

Protein-protein interaction studies using proximity labeling techniques, such as biotin ligase-based BioID, have become integral in understanding cellular processes. Most studies utilize conventional 2D cell culture systems, potentially missing important differences in protein behavior found in 3D tissues. In this study, we investigated the protein-protein interactions of a protein, Bcl-2 Agonist of cell death (BAD), and compared conventional 2D culture conditions to a 3D system, wherein cells were embedded within a 3D extracellular matrix (ECM) mimic. Using BAD fused to the engineered biotin ligase miniTurbo (BirA*), we identified both overlapping and distinct BAD interactomes under 2D and 3D conditions. The known BAD binding proteins 14-3-3 isoforms and Bcl-XL interacted with BAD in both 2D and 3D. Of the 131 BAD-interactors identified, 56% were specific to 2D, 14% were specific to 3D, and 30% were common to both conditions. Interaction network analysis demonstrated differential associations between 2D and 3D interactomes, emphasizing the impact of the culture conditions on protein interactions. The 2D-3D overlap interactome encapsulated the apoptotic program, which is a well-known role of BAD. The 3D unique pathways were enriched in ECM signaling, suggestive of hitherto unknown functions for BAD. Thus, exploring protein-protein interactions in 3D provides novel clues into cell behavior. This exciting approach has the potential to bridge the knowledge gap between tractable 2D cell culture and organoid-like 3D systems.