基于机器学习的药物-靶标相互作用预测*

近年来,随着计算机硬件、软件工具和数据丰度的不断突破,以机器学习为代表的人工智能技术在生物、基础医学和药学等领域的应用不断拓展和融合,极大地推动了这些领域的发展,尤其是药物研发领域的变革。其中,药物-靶标相互作用(drug-target interactions, DTI)的识别是药物研发领域中的重要难题和人工智能技术交叉融合的热门方向,研究人员在DTI预测方面做了大量的工作,构建了许多重要的数据库,开发或拓展了各类机器学习算法和工具软件。对基于机器学习的DTI预测的基本流程进行了介绍,并对利用机器学习预测DTI的研究进行了回顾,同时对不同的机器学习方法运用于DTI预测的优缺点进行了简单总结,以期对开发更加有效的预测算法和DTI预测的发展提供帮助。     

Alterations to plasma membrane lipid contents affect the biophysical properties of erythrocytes from individuals with hypertension

Genetic and environmental factors may contribute to high blood pressure, which is termed essential hypertension. Hypertension is a major independent risk factor for cardiovascular disease, stroke and renal failure; thus, elucidation of the etiopathology of hypertension merits further research. We recently reported that the platelets and neutrophils of patients with hypertension exhibit altered biophysical characteristics. In the present study, we assessed whether the major structural elements of erythrocyte plasma membranes are altered in individuals with hypertension. We compared the phospholipid (phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, sphingosine) and cholesterol contents of erythrocytes from individuals with hypertension (HTN) and healthy individuals (HI) using LC/MS-MS. HTN erythrocytes contained higher phosphatidylcholine, phosphatidylethanolamine and phosphatidylserine contents and a lower cholesterol content than HI erythrocytes. Furthermore, atomic force microscopy revealed important morphological changes in HTN erythrocytes, which reflected the increased membrane fragility and fluidity and higher levels of oxidative stress observed in HTN erythrocytes using spectrophotofluorometry, flow cytometry and spectrometry. This study reveals that alterations to the lipid contents of erythrocyte plasma membranes occur in hypertension, and these alterations in lipid composition result in morphological and physiological abnormalities that modify the dynamic properties of erythrocytes and contribute to the pathophysiology of hypertension.     

Deconstructing membrane traffic

Intracellular movement of proteins and lipids between organelles is usually described in terms of cargo, carriers, traffic and docking, familiar terms that imply parallels to human activities. Over the past century, scientists have been criticized for constructing hypotheses that reflect too much of their current political and cultural values. In this article, concepts of membrane traffic are re-examined to see whether they reflect the cell’s view of the world or our own.     

A new hybrid approach to predict subcellular localization of proteins by incorporating gene ontology

Based on the recent development in the gene ontology and functional domain databases, a new hybridization approach is developed for predicting protein subcellular location by combining the gene product, functional domain, and quasi-sequence-order effects. As a showcase, the same prokaryotic and eukaryotic datasets, which were studied by many previous investigators, are used for demonstration. The overall success rate by the jackknife test for the prokaryotic set is 94.7% and that for the eukaryotic set 92.9%. These are so far the highest success rates achieved for the two datasets by following a rigorous cross-validation test procedure, suggesting that such a hybrid approach may become a very useful high-throughput tool in the area of bioinformatics, proteomics, as well as molecular cell biology. The very high success rates also reflect the fact that the subcellular localization of a protein is closely correlated with: (1). the biological objective to which the gene or gene product contributes, (2). the biochemical activity of a gene product, and (3). the place in the cell where a gene product is active.     

On the minimum number of topologies explaining a sample of DNA sequences

In this article I derive an alternative algorithm to Hudson and Kaplan's (Genetics 111, 147-165) algorithm that gives a lower bound to the number of recombination events in a sample's history. It is shown that the number, T(M), found by the algorithm is the least number of topologies required to explain a set of DNA sequences sampled under the infinite-site assumption. Let Tao = (T(1),...,T(r)) be a list of topologies compatible with the sequences, i.e., T(k) is compatible with an interval, I(k), of sites in the alignment. A characterization of all lists having T(M) topologies is given and it is shown that T(M) relates to specific patterns in the alignment, here called chain series. Further, a number of theorems relating general lists of topologies to the number T(M) is presented. The results are discussed in relation to the true minimum number of recombination events required to explain an alignment.     

Sphingolipid trafficking and protein sorting in epithelial cells

Sphingolipids represent a minor, but highly dynamic subclass of lipids in all eukaryotic cells. They are involved in functions that range from structural protection to signal transduction and protein sorting, and participate in lipid raft assembly. In polarized epithelial cells, which display an asymmetric apical and basolateral membrane surface, rafts have been proposed as a sorting principle for apical resident proteins, following their biosynthesis. However, raft-mediated trafficking is ubiquitous in cells. Also, sphingolipids per se, which are strongly enriched in the apical domain, are subject to sorting in polarity development. Next to the trans Golgi network, a subapical compartment called SAC or common endosome appears instrumental in regulating these sorting events.     

Exosome and Exosomal MicroRNA:Trafficking,Sorting, and Function

Exosomes are 40–100 nm nano-sized vesicles that are released from many cell types into the extracellular space. Such vesicles are widely distributed in various body fluids. Recently,m RNAs and micro RNAs(mi RNAs) have been identified in exosomes, which can be taken up by neighboring or distant cells and subsequently modulate recipient cells. This suggests an active sorting mechanism of exosomal mi RNAs, since the mi RNA profiles of exosomes may differ from those of the parent cells. Exosomal mi RNAs play an important role in disease progression, and can stimulate angiogenesis and facilitate metastasis in cancers. In this review, we will introduce the origin and the trafficking of exosomes between cells, display current research on the sorting mechanism of exosomal mi RNAs, and briefly describe how exosomes and their mi RNAs function in recipient cells.Finally, we will discuss the potential applications of these mi RNA-containing vesicles in clinical settings.     

rCNT2 extracellular cysteines,Cys615 and Cys649, are important for maturation and sorting to the plasma membrane

rCNT2 is a purine-preferring concentrative nucleoside transporter implicated in the regulation of extracellular adenosine levels and purinergic signaling. This study addressed the analysis of the CNT2 C-terminus tail as a domain likely to be implicated in transporter sorting. The topological mapping of this segment revealed that Cys⁶¹⁵ and Cys⁶⁴⁹ are important residues for the proper trafficking of CNT2 to the plasma membrane. The inhibition of protein disulfide isomerase (PDI) and ER glycosidase I and II impaired rCNT2 trafficking to the cell surface, similarly to Cys⁶¹⁵ and Cys⁶⁴⁹ mutants. The present work suggests these two cysteine residues are relevant for the proper sorting of the transporter and its functional performance.     

免疫进化算法在建立地位指数曲线模型中的应用

选择Schumacher方程构建优势高和地位指数均能写成显式表达式的多形地位指数曲线模型,以同时兼顾优势高和地位指数的估测误差最小为目标函数,采用免疫进化算法求解参数．实例分析表明：该文建立的多形地位指数曲线模型能够客观地反映不同立地下优势高的生长规律,且误差小,应用方便．