A protein sequence-based deep transfer learning framework for identifying human proteome-wide deubiquitinase-substrate interactions.

Protein ubiquitination regulates a wide range of cellular processes. The degree of protein ubiquitination is determined by the delicate balance between ubiquitin ligase (E3)-mediated ubiquitination and deubiquitinase (DUB)-mediated deubiquitination. In comparison to the E3-substrate interactions, the DUB-substrate interactions (DSIs) remain insufficiently investigated. To address this challenge, we introduce a protein sequence-based ab initio method, TransDSI, which transfers proteome-scale evolutionary information to predict unknown DSIs despite inadequate training datasets. An explainable module is integrated to suggest the critical protein regions for DSIs while predicting DSIs. TransDSI outperforms multiple machine learning strategies against both cross-validation and independent test. Two predicted DUBs (USP11 and USP20) for FOXP3 are validated by "wet lab" experiments, along with two predicted substrates (AR and p53) for USP22. TransDSI provides new functional perspective on proteins by identifying regulatory DSIs, and offers clues for potential tumor drug target discovery and precision drug application.

An integrated computational strategy to predict personalized cancer drug combinations by reversing drug resistance signatures.

Drug resistance currently poses the greatest barrier to cancer treatments. To overcome drug resistance, drug combination therapy has been proposed as a promising treatment strategy. Herein, we present Re-Sensitizing Drug Prediction (RSDP), a novel computational strategy, for predicting the personalized cancer drug combination A + B by reversing the resistance signature of drug A. The process integrates multiple biological features using a robust rank aggregation algorithm, including Connectivity Map, synthetic lethality, synthetic rescue, pathway, and drug target. Bioinformatics assessments revealed that RSDP achieved a relatively accurate prediction performance for identifying personalized combinational re-sensitizing drug B against cell line-specific intrinsic resistance, cell line-specific acquired resistance, and patient-specific intrinsic resistance to drug A. In addition, we developed the largest resource of cell line-specific cancer drug resistance signatures, including intrinsic and acquired resistance, as a byproduct of the proposed strategy. The findings indicate that personalized drug resistance signature reversal is a promising strategy for identifying personalized drug combinations, which may guide future clinical decisions regarding personalized medicine.

Research progress of chlorogenic acid in improving inflammatory diseases. / ç»¿åŽŸé ¸æ”¹å–„ç‚Žç—‡æ€§ç–¾ç— çš„ç ”ç©¶è¿›å±•.

Long-term inflammation will develop into chronic inflammation and become inflammatory diseases. Antibiotics are commonly used in clinical practice to treat inflammatory diseases. But patients are prone to drug resistance. So we need to find new treatment. Chlorogenic acid is an organic compound extracted from honeysuckle and other plants. Its anti-inflammatory activity is strong, and it has a significant anti-inflammatory effect on inflammatory diseases in various systems. It has been shown that chlorogenic acid can regulate inflammation-related signaling pathways, such as nuclear factor κB (NF-κB) canonical signaling pathway, NF-κB atypical signaling pathway, nuclear factor-erythroid 2-related factor 2 (Nrf2) canonical signaling pathway, and Nrf2 atypical signaling pathway, etc. It can up-regulate the expression of anti-inflammatory cytokines such as interleukin (IL)-4, IL-10, IL-13 and down-regulate the expression of pro-inflammatory cytokine such as IL-1ß, IL-6, and IL-8. Although chlorogenic acid has a strong anti-inflammatory effect, but clinical trials and application still face many difficulties. In the future, the anti-inflammatory molecular mechanism of chlorogenic acid should be further studied to explore its clinical application value and improve new ideas for the treatment of inflammatory diseases.

Integration of Human Protein Sequence and Protein-Protein Interaction Data by Graph Autoencoder to Identify Novel Protein-Abnormal Phenotype Associations.

Understanding gene functions and their associated abnormal phenotypes is crucial in the prevention, diagnosis and treatment against diseases. The Human Phenotype Ontology (HPO) is a standardized vocabulary for describing the phenotype abnormalities associated with human diseases. However, the current HPO annotations are far from completion, and only a small fraction of human protein-coding genes has HPO annotations. Thus, it is necessary to predict protein-phenotype associations using computational methods. Protein sequences can indicate the structure and function of the proteins, and interacting proteins are more likely to have same function. It is promising to integrate these features for predicting HPO annotations of human protein. We developed GraphPheno, a semi-supervised method based on graph autoencoders, which does not require feature engineering to capture deep features from protein sequences, while also taking into account the topological properties in the protein-protein interaction network to predict the relationships between human genes/proteins and abnormal phenotypes. Cross validation and independent dataset tests show that GraphPheno has satisfactory prediction performance. The algorithm is further confirmed on automatic HPO annotation for no-knowledge proteins under the benchmark of the second Critical Assessment of Functional Annotation, 2013-2014 (CAFA2), where GraphPheno surpasses most existing methods. Further bioinformatics analysis shows that predicted certain phenotype-associated genes using GraphPheno share similar biological properties with known ones. In a case study on the phenotype of abnormality of mitochondrial respiratory chain, top prioritized genes are validated by recent papers. We believe that GraphPheno will help to reveal more associations between genes and phenotypes, and contribute to the discovery of drug targets.

UbiBrowser 2.0: a comprehensive resource for proteome-wide known and predicted ubiquitin ligase/deubiquitinase-substrate interactions in eukaryotic species.

As an important post-translational modification, ubiquitination mediates ∼80% of protein degradation in eukaryotes. The degree of protein ubiquitination is tightly determined by the delicate balance between specific ubiquitin ligase (E3)-mediated ubiquitination and deubiquitinase-mediated deubiquitination. In 2017, we developed UbiBrowser 1.0, which is an integrated database for predicted human proteome-wide E3-substrate interactions. Here, to meet the urgent requirement of proteome-wide E3/deubiquitinase-substrate interactions (ESIs/DSIs) in multiple organisms, we updated UbiBrowser to version 2.0 (http://ubibrowser.ncpsb.org.cn). Using an improved protocol, we collected 4068/967 known ESIs/DSIs by manual curation, and we predicted about 2.2 million highly confident ESIs/DSIs in 39 organisms, with >210-fold increase in total data volume. In addition, we made several new features in the updated version: (i) it allows exploring proteins' upstream E3 ligases and deubiquitinases simultaneously; (ii) it has significantly increased species coverage; (iii) it presents a uniform confidence scoring system to rank predicted ESIs/DSIs. To facilitate the usage of UbiBrowser 2.0, we also redesigned the web interface for exploring these known and predicted ESIs/DSIs, and added functions of 'Browse', 'Download' and 'Application Programming Interface'. We believe that UbiBrowser 2.0, as a discovery tool, will contribute to the study of protein ubiquitination and the development of drug targets for complex diseases.