Protein and Peptide Letters - Volume 27, Issue 5, 2020

The identification of Drug-Target Interactions (DTIs) is an important process in drug discovery and medical research. However, the tradition experimental methods for DTIs identification are still time consuming, extremely expensive and challenging. In the past ten years, various computational methods have been developed to identify potential DTIs. In this paper, the identification methods of DTIs are summarized. What's more, several state-of-the-art computational methods are mainly introduced, containing network-based method and machine learning-based method. In particular, for machine learning-based methods, including the supervised and semisupervised models, have essential differences in the approach of negative samples. Although these effective computational models in identification of DTIs have achieved significant improvements, network-based and machine learning-based methods have their disadvantages, respectively. These computational methods are evaluated on four benchmark data sets via values of Area Under the Precision Recall curve (AUPR).

Protein-related interaction prediction is critical to understanding life processes, biological functions, and mechanisms of drug action. Experimental methods used to determine proteinrelated interactions have always been costly and inefficient. In recent years, advances in biological and medical technology have provided us with explosive biological and physiological data, and deep learning-based algorithms have shown great promise in extracting features and learning patterns from complex data. At present, deep learning in protein research has emerged. In this review, we provide an introductory overview of the deep neural network theory and its unique properties. Mainly focused on the application of this technology in protein-related interactions prediction over the past five years, including protein-protein interactions prediction, protein-RNA\DNA, Protein– drug interactions prediction, and others. Finally, we discuss some of the challenges that deep learning currently faces.

Understanding the interaction mechanism of proteins and nucleic acids is one of the most fundamental problems for genome editing with engineered nucleases. Due to some limitations of experimental investigations, computational methods have played an important role in obtaining the knowledge of protein-nucleic acid interaction. Over the past few years, dozens of computational tools have been used for identification of nucleic acid binding site for site-specific proteins and design of site-specific nucleases because of their significant advantages in genome editing. Here, we review existing widely-used computational tools for target prediction of site-specific proteins as well as off-target prediction of site-specific nucleases. This article provides a list of on-line prediction tools according to their features followed by the description of computational methods used by these tools, which range from various sequence mapping algorithms (like Bowtie, FetchGWI and BLAST) to different machine learning methods (such as Support Vector Machine, hidden Markov models, Random Forest, elastic network and deep neural networks). We also make suggestions on the further development in improving the accuracy of prediction methods. This survey will provide a reference guide for computational biologists working in the field of genome editing.

In recent years, more and more evidence indicates that long non-coding RNA (lncRNA) plays a significant role in the development of complex biological processes, especially in RNA progressing, chromatin modification, and cell differentiation, as well as many other processes. Surprisingly, lncRNA has an inseparable relationship with human diseases such as cancer. Therefore, only by knowing more about the function of lncRNA can we better solve the problems of human diseases. However, lncRNAs need to bind to proteins to perform their biomedical functions. So we can reveal the lncRNA function by studying the relationship between lncRNA and protein. But due to the limitations of traditional experiments, researchers often use computational prediction models to predict lncRNA protein interactions. In this review, we summarize several computational models of the lncRNA protein interactions prediction base on semi-supervised learning during the past two years, and introduce their advantages and shortcomings briefly. Finally, the future research directions of lncRNA protein interaction prediction are pointed out.

Self-Interacting Proteins (SIPs), whose two or more copies can interact with each other, have significant roles in cellular functions and evolution of Protein Interaction Networks (PINs). Knowing whether a protein can act on itself is important to understand its functions. Previous studies on SIPs have focused on their structures and functions, while their whole properties are less emphasized. Not surprisingly, identifying SIPs is one of the most important works in biomedical research, which will help to understanding the function and mechanism of proteins. It is worth noting that high throughput methods can be used for SIPs prediction, but can be costly, time consuming and challenging. Therefore, it is urgent to design computational models for the identification of SIPs. In this review, the concept and function of SIPs were introduced in detail. We further introduced SIPs data and some excellent computational models that have been designed for SIPs prediction. Specially, the most existing approaches were developed based on machine learning through carrying out different extract feature methods. Finally, we discussed several difficult problems in developing computational models for SIPs prediction.

Enzymes are among the most studied biological molecules because better understanding enzymes structure and activity will shed more light on their biological processes and regulation; from a biotechnological point of view there are many examples of enzymes used with the aim to obtain new products and/or to make industrial processes less invasive towards the environment. Enzymes are known for their high specificity in the recognition of a substrate but considering the particular features of an increasing number of enzymes this is not completely true, in fact, many enzymes are active on different substrates: this ability is called enzyme promiscuity. Usually, promiscuous activities have significantly lower kinetic parameters than to that of primary activity, but they have a crucial role in gene evolution. It is accepted that gene duplication followed by sequence divergence is considered a key evolutionary mechanism to generate new enzyme functions. In this way, promiscuous activities are the starting point to increase a secondary activity in the main activity and then get a new enzyme. The primary activity can be lost or reduced to a promiscuous activity. In this review we describe the differences between substrate and enzyme promiscuity, and its rule in gene evolution. From a practical point of view the knowledge of promiscuity can facilitate the in vitro progress of proteins engineering, both for biomedical and industrial applications. In particular, we report cases regarding esterases, phosphotriesterases and cytochrome P450.

Background: microRNAs (miRNAs) are a small, endogenous non-coding RNAs that are involved in post-transcriptional gene regulation of many biological processes, including embryo implantation and placental development. In our previous study, miR-146a-5p was found expressed higher in the serum exosomes of pregnant sows than non-pregnant. The research on miR-146a-5p has been mainly related to human diseases, but there are few studies on its effects on the reproduction of sows in early pregnancy. Objective: In this article, our motivation is to study the role of miR-146a-5p in the early pregnancy of sows on the cell proliferetion and apoptosis by targeting SMAD3 and SMAD4. Methods: Bioinformatics software was used to identify the target genes of miR-146a-5p. The wildtype and mutant-type recombinant plasmids of dual-luciferase reporter with 3'-UTR of Smad3 or 3'- UTR of Smad4 were constructed, and co-transfected in porcine kidney cell (PK-15 cell) with miR- 146a-5p mimic, mimic-NC(M-NC), inhibitor and inhibitor-NC(IN-NC), then dual-luciferase activity analysis, qRT-PCR and Western blot were performed to verify the target genes. After the transfection of BeWo choriocarcinoma cell (BeWo cell) with miR-146a-5p mimic, M-NC, inhibitor and IN-NC, the mRNA expression of Caspase-3, BAX and Bcl-2 was measured using qRT-PCR, and the cell proliferation was measured using CCK-8 kit. Results: The luciferase, mRNA and protein expression of Smad3 in PK-15 cells treated by Smad3- 3'-UTR-W co-transfected with miR-146a-5p mimic were significantly lower than that with miR- 146a-5p M-NC, and the results of Smad4 were similar to Smad3, but the protein expression had a trend to lower in mimic group. The expression level of Bcl-2 in the miR-146a-5p mimic group was significantly lower than that in the miR-146a-5p M-NC group, but the expression pattern of Caspase-3 was just opposite. The mimic of miR-146a-5p reduced the proliferation of BeWo cells, however the inhibitor increased. Conclusion: Smad3 and Smad4 are the direct target genes of miR-146a-5p. The expression of Smad3 and Smad4 were affected by the mimic and inhibitor of miR-146a-5p. miR-146a-5p affects cell apoptosis and proliferation by regulating their target genes. This study provided new data to understand the regulation mechanism of early pregnancy in sows.

Background: Insulin is a peptide hormone used for regulating blood glucose levels. Human insulin market is projected to grow at a rate of 12.5% annually. To meet the needs of patients, a cost effective insulin manufacturing strategy has to be developed. This can be achieved by selecting a competent host, ideal fusion tag and streamlined downstream process. Objective: In this article, we have demonstrated that selecting a right fusion partner for expression of toxic proteins like insulin, plays a major role in increasing the recombinant protein yield. Methods: In this article, we have focused on identifying a peptide tag fusion partner for expressing proinsulin by truncating thioredoxin tag. Truncations were carried out from both Amino and Carboxy terminus of the protein and efficiency of truncated sequences was evaluated by expressing it with proinsulin gene. FCTRX (1-15) sequence fused to proinsulin was processed further to establish downstream protocol for purification. Results: Thioredoxin tag was truncated appropriately by considering the fusion tag: protein ratio. A couple of sequences ranging 10 – 15 amino acids were identified based on its in silico properties. Of these FCTRX (1-15) showed increased expression and stability of fusion protein. 156 mg of purified insulin was generated from 1g of inclusion body after enzymatic conversion and chromatographic steps. Conclusion: As a result of the current study, it was concluded that FCTRX (1-15) peptide has advantageous attributes to be considered as an ideal fusion tag for expression of proinsulin. This can be further explored by expressing it with other proteins.

Background: Intermediate Filaments (IFs) are major constituents of the cytoskeletal systems in animal cells. Objective: To gain insights into the structure-function relationship of invertebrate cytoplasmic IF proteins, we characterized an IF protein from the platyhelminth, Dugesia japonica, termed Dif-1. Methods: cDNA cloning, in situ hybridization, immunohistochemical analysis, and IF assembly experiments in vitro using recombinant Dif-1, were performed for protein characterization. Results: The structure deduced from the cDNA sequence showed that Djf-1 comprises 568 amino acids and has a tripartite domain structure (N-terminal head, central rod, and C-terminal tail) that is characteristic of IF proteins. Similar to nuclear IF lamins, Djf-1 contains an extra 42 residues in the coil 1b subdomain of the rod domain that is absent from vertebrate cytoplasmic IF proteins and a nuclear lamin-homology segment of approximately 105 residues in the tail domain; however, it contains no nuclear localization signal. In situ hybridization analysis showed that Djf-1 mRNA is specifically expressed in cells located within the marginal region encircling the worm body. Immunohistochemical analysis showed that Djf-1 protein forms cytoplasmic IFs located close to the microvilli of the cells. In vitro IF assembly experiments using recombinant proteins showed that Djf-1 alone polymerizes into IFs. Deletion of the extra 42 residues in the coil 1b subdomain resulted in the failure of IF formation. Conclusion: Together with data from other histological studies, our results suggest that Djf- 1 is expressed specifically in anchor cells within the glandular adhesive organs of the worm and that Djf-1 IFs may play a role in protecting the cells from mechanical stress.