人源microRNA前体的全基因组预测

microRNA（miRNA）是一类不编码蛋白的调控小分子RNA,在真核生物中发挥着广泛而重要的调控功能.由于miRNA的表达具有时空特异性,因而通过计算方法预测miRNA而后有针对性的实验验证是miRNA发现的一条重要途径.降低假阳性率是miRNA预测方法面临的重要挑战.本研究采用集成学习方法构建预测miRNA前体的分类器SVMbagging,对训练集、测试集和独立测试集的结果表明,本研究的方法性能稳健、假阳性率低,具有很好的泛化能力,尤其是当阈值取0.9时,特异性高达99.90%,敏感性在26%以上,适合于全基因组预测.采用SVMbagging在人全基因组中预测miRNA前体,当取阈值0.9时,得到14933个可能的miRNA前体.通过与高通量小RNA测序数据的比较,发现其中4481个miRNA前体具有完全匹配的小RNA序列,与理论估计的真阳性数值非常接近.最后,对32个可能的miRNA进行实验验证,确定其中2条为真实的miRNA.     

iSulf_wide-PseAAC: 基于集成支持向量机的蛋白质S-亚磺酰化预测方法

蛋白质S-亚磺酰化是可逆的蛋白质翻译后修饰(post-translational modifications, PTMs),在生物生长中发挥至关重要的作用。同时,它与一些疾病相关。因此,无论是从基础研究还是药物开发的角度,都面临着一个具有挑战性的问题:哪些是属于S-亚磺酰化位点？为了解决这个问题,本文开发了一种基于机器学习的预测方法。该系统主要步骤为:(1)将这些蛋白质组合成等长度的伪氨基酸;(2)使用下采样方法来平衡训练数据集;(3)通过集成方法建立一个综合的预测系统进行预测。最终,得到的独立测试集的准确率达到90.77％,其他各个指标对比现有方法提升效果明显,为生物信息学的发展提供了帮助。本文建立了一个友好的web服务器预测网站:http: //www.jci-bioinfo.cn/iSulf_Wide-PseAAC,通过该网站不需要复杂的计算公式即可在线预测,它将为用户提供便利和进一步研究的指南。与此同时,本文中使用到的数学方法会解决类似相关领域的诸多其他问题。     

iSucc-PseAAC:基于集成机器学习的赖氨酸琥珀酰化修饰位点预测

赖氨酸琥珀酰化是一种新型的翻译后修饰,在蛋白质调节和细胞功能控制中发挥重要作用,所以准确识别蛋白质中的琥珀酰化位点是有必要的。传统的实验耗费物力和财力。通过计算方法预测是近段时间以来提出的一种高效的预测方法。本研究中,我们开发了一种新的预测方法iSucc-PseAAC,它是通过使用多种分类算法结合不同的特征提取方法。最终发现,基于耦合序列(PseAAC)特征提取下,使用支持向量机分类效果是最好的,并结合集成学习解决了数据不平衡问题。与现有方法预测效果对比,iSucc-PseAAC在区分赖氨酸琥珀酰化位点方面,更具有意义和实用性。     

Substituting random forest for multiple linear regression improves binding affinity prediction of scoring functions: Cyscore as a case study

Background

State-of-the-art protein-ligand docking methods are generally limited by the traditionally low accuracy of their scoring functions, which are used to predict binding affinity and thus vital for discriminating between active and inactive compounds. Despite intensive research over the years, classical scoring functions have reached a plateau in their predictive performance. These assume a predetermined additive functional form for some sophisticated numerical features, and use standard multivariate linear regression (MLR) on experimental data to derive the coefficients.

Results

In this study we show that such a simple functional form is detrimental for the prediction performance of a scoring function, and replacing linear regression by machine learning techniques like random forest (RF) can improve prediction performance. We investigate the conditions of applying RF under various contexts and find that given sufficient training samples RF manages to comprehensively capture the non-linearity between structural features and measured binding affinities. Incorporating more structural features and training with more samples can both boost RF performance. In addition, we analyze the importance of structural features to binding affinity prediction using the RF variable importance tool. Lastly, we use Cyscore, a top performing empirical scoring function, as a baseline for comparison study.

Conclusions

Machine-learning scoring functions are fundamentally different from classical scoring functions because the former circumvents the fixed functional form relating structural features with binding affinities. RF, but not MLR, can effectively exploit more structural features and more training samples, leading to higher prediction performance. The future availability of more X-ray crystal structures will further widen the performance gap between RF-based and MLR-based scoring functions. This further stresses the importance of substituting RF for MLR in scoring function development.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2105-15-291) contains supplementary material, which is available to authorized users.     

Structure-based Discovery of Novel Small Molecule Wnt Signaling Inhibitors by Targeting the Cysteine-rich Domain of Frizzled

Frizzled is the earliest discovered glycosylated Wnt protein receptor and is critical for the initiation of Wnt signaling. Antagonizing Frizzled is effective in inhibiting the growth of multiple tumor types. The extracellular N terminus of Frizzled contains a conserved cysteine-rich domain that directly interacts with Wnt ligands. Structure-based virtual screening and cell-based assays were used to identify five small molecules that can inhibit canonical Wnt signaling and have low IC₅₀ values in the micromolar range. NMR experiments confirmed that these compounds specifically bind to the Wnt binding site on the Frizzled8 cysteine-rich domain with submicromolar dissociation constants. Our study confirms the feasibility of targeting the Frizzled cysteine-rich domain as an effective way of regulating canonical Wnt signaling. These small molecules can be further optimized into more potent therapeutic agents for regulating abnormal Wnt signaling by targeting Frizzled.     

Identifying the novel inhibitors of lysine-specific demethylase 1 (LSD1) combining pharmacophore-based and structure-based virtual screening

Abstract

Lysine-specific demethylase 1 (LSD1) has been reported to connect with a range of solid tumors. Thus, the exploration of LSD1 inhibitors has emerged as an effective strategy for cancer treatment. In this study, we constructed a pharmacophore model based on a series of flavin adenine dinucleotide (FAD)-competing inhibitors bearing triazole???dithiocarbamate scaffold combining docking, structure–activity relationship (SAR) study, and molecular dynamic (MD) simulation. Meanwhile, another pharmacophore model was also constructed manually, relying on several speculated substrate-competing inhibitors and reported putative vital interactions with LSD1. On the basis of the two pharmacophore models, multi-step virtual screenings (VSs) were performed against substrate-binding pocket and FAD-binding pocket, respectively, combining pharmacophore-based and structure-based strategy to exploit novel LSD1 inhibitors. After bioassay evaluation, four compounds among 21 hits with diverse and novel scaffolds exhibited inhibition activity at the range of 3.63–101.43?μM. Furthermore, substructure-based enrichment was performed, and four compounds with a more potent activity were identified. After that, the time-dependent assay proved that the most potent compound with IC₅₀ 2.21?μM inhibits LSD1 activity in a manner of time-independent. In addition, the compound exhibited a cellular inhibitory effect against LSD1 in MGC-803 cells and may inhibit cell migration and invasion by reversing EMT in cultured gastric cancer cells. Considering the binding mode and SAR of the series of compounds, we could roughly deem that these compounds containing 3-methylxanthine scaffold act through occupying substrate-binding pocket competitively. This study presented a new starting point to develop novel LSD1 inhibitors.