Dear Colleagues,

Proteins are the direct and major performer for biological processes and molecular functions in cells. Protein–protein interactions (PPIs) are generally defined as the physical interactions between two or more protein molecules, endogenous or exogenous. Comprehensively, PPIs can be divided into two major groups: 1) stable PPIs, which mainly help to establish a common effective protein complex; and 2) transient PPIs, which mainly help to mediate biological processes that are essential for living cells. Interactions between two or more proteins in living cells are one of the most essential regulatory approaches for basic cell biological functions including movement, reproduction, stimulation response, nutrition, excretion, respiration, and growth. Therefore, studies on PPIs are the most direct and effective way to reveal the complex regulatory mechanisms in cells. Abnormal PPIs have been observed during the cellular pathogenic transformation. Associations between abnormal PPIs and pathogenesis have been validated by multiple experimental and computational methods. Therefore, PPI recognition can also help us to find out disease-specific biomarkers, develop novel target drugs, and reveal the pathogenesis of diseases.

Current studies on PPIs can be divided into two major aspects:

1) Experimental-based PPI screening;

2) Computational methods based on PPI predictions.

Experimental-based PPI screening is the most classic and reliable method to identify and validate protein–protein interactions. However, even up to today, most experimental solutions for PPI recognition rely on expensive and time-consuming methods such as co-immunoprecipitation (Co-IP), pull-down assays, nuclear magnetic resonance spectroscopy (NMR), X-ray crystallography, electron microscopy, label transfer protein interaction analyses, and far-Western blot analysis, especially for high-throughput screening procedures.

Computational methods for PPI prediction apply computational algorithms such as machine learning models, deep learning models, etc., on archived databases or experimental results to predict novel protein–protein interactions. Different from experimental approaches, computational methods can focus on the amino acid sequence of the protein motif, protein topological structures, or the existing protein–protein interaction networks to predict novel protein–protein interactions and explore their potential biological significance. PPI prediction using computational methods has two major requirements: 1) proficient analyzable data; and 2) effective and producible statistical methods. In terms of analyzable data for PPIs, in recent years, high-throughput technologies and accumulated proteomics experimental results have collected massive omics data including proteomics, transcriptomics, and interactomics datasets for PPI screening and upstream/downstream functional exploration. Methodologically, more and more statistical algorithms including machine learning models such as Support Vector Machine, Random Forest, Random Walk with Restart, and deep learning models such as Convolutional Neural Networks have been applied to optimize the feature encoding or prediction model establishment procedures and recognize potential disease-associated protein–protein interactions.

However, several limitations and challenges remain for computational methods-based PPI analyses:

1) The general accuracy and efficacy for protein–protein interaction are still limited and massive potential disease-associated PPIs remain unrecognized. More updated algorithms should be introduced to identify disease-associated protein–protein interactions and build up comprehensive protein–protein interaction networks under physiological or pathological conditions.

2) Few studies try to recognize the upstream and downstream regulatory mechanisms for PPIs. Extensive analyses on interactions between protein complexes (stable PPIs) and other molecules such as RNA, small molecule drugs, and metabolites may help validate the significance of protein–protein interactions in living cells.

3) Systematic comparisons between multiple computational models for PPI prediction can help us identify the optimized model for disease-associated PPI prediction.

Therefore, in this research topic, we focused on applying various machine learning models, deep learning models, or network analyses to recognize potential protein–protein interactions, identify optimized protein–protein interaction prediction models, and build up novel protein–protein interaction networks under physiological and pathological conditions. Welcomed study designs are listed below:

1) Studies identifying novel disease-associated protein–protein interactions using machine learning or deep learning models;

2) Studies exploring biological implications for significant protein–protein interactions or connecting protein–protein interactions with other upstream (genomics, epigenomics, and transcriptomics) or downstream (metabolomics) omics regulators;

3) Studies comparing multiple computational methods to identify the optimized protein–protein interactions.

Dr. Yu-Hang Zhang
Dr. Guohua Huang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Life is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• protein–protein interactions
• machine learning
• deep learning
• disease biomarkers
• network analyses
• interactome