CRCS: An automatic image processing pipeline for hormone level analysis of Cushing's disease.

Due to the abnormal secretion of adreno-cortico-tropic-hormone (ACTH) by tumors, Cushing's disease leads to hypercortisonemia, a precursor to a series of metabolic disorders and serious complications. Cushing's disease has high recurrence rate, short recurrence time and undiscovered recurrence reason after surgical resection. Qualitative or quantitative automatic image analysis of histology images can potentially in providing insights into Cushing's disease, but still no software has been available to the best of our knowledge. In this study, we propose a quantitative image analysis-based pipeline CRCS, which aims to explore the relationship between the expression level of ACTH in normal cell tissues adjacent to tumor cells and the postoperative prognosis of patients. CRCS mainly consists of image-level clustering, cluster-level multi-modal image registration, patch-level image classification and pixel-level image segmentation on the whole slide imaging (WSI). On both image registration and classification tasks, our method CRCS achieves state-of-the-art performance compared to recently published methods on our collected benchmark dataset. In addition, CRCS achieves an accuracy of 0.83 for postoperative prognosis of 12 cases. CRCS demonstrates great potential for instrumenting automatic diagnosis and treatment for Cushing's disease.

GraphLoc: a graph neural network model for predicting protein subcellular localization from immunohistochemistry images.

MOTIVATION: Recognition of protein subcellular distribution patterns and identification of location biomarker proteins in cancer tissues are important for understanding protein functions and related diseases. Immunohistochemical (IHC) images enable visualizing the distribution of proteins at the tissue level, providing an important resource for the protein localization studies. In the past decades, several image-based protein subcellular location prediction methods have been developed, but the prediction accuracies still have much space to improve due to the complexity of protein patterns resulting from multi-label proteins and the variation of location patterns across cell types or states. RESULTS: Here, we propose a multi-label multi-instance model based on deep graph convolutional neural networks, GraphLoc, to recognize protein subcellular location patterns. GraphLoc builds a graph of multiple IHC images for one protein, learns protein-level representations by graph convolutions and predicts multi-label information by a dynamic threshold method. Our results show that GraphLoc is a promising model for image-based protein subcellular location prediction with model interpretability. Furthermore, we apply GraphLoc to the identification of candidate location biomarkers and potential members for protein networks. A large portion of the predicted results have supporting evidence from the existing literatures and the new candidates also provide guidance for further experimental screening. AVAILABILITY AND IMPLEMENTATION: The dataset and code are available at: www.csbio.sjtu.edu.cn/bioinf/GraphLoc. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MDGF-MCEC: a multi-view dual attention embedding model with cooperative ensemble learning for CircRNA-disease association prediction.

Circular RNA (circRNA) is closely involved in physiological and pathological processes of many diseases. Discovering the associations between circRNAs and diseases is of great significance. Due to the high-cost to verify the circRNA-disease associations by wet-lab experiments, computational approaches for predicting the associations become a promising research direction. In this paper, we propose a method, MDGF-MCEC, based on multi-view dual attention graph convolution network (GCN) with cooperative ensemble learning to predict circRNA-disease associations. First, MDGF-MCEC constructs two disease relation graphs and two circRNA relation graphs based on different similarities. Then, the relation graphs are fed into a multi-view GCN for representation learning. In order to learn high discriminative features, a dual-attention mechanism is introduced to adjust the contribution weights, at both channel level and spatial level, of different features. Based on the learned embedding features of diseases and circRNAs, nine different feature combinations between diseases and circRNAs are treated as new multi-view data. Finally, we construct a multi-view cooperative ensemble classifier to predict the associations between circRNAs and diseases. Experiments conducted on the CircR2Disease database demonstrate that the proposed MDGF-MCEC model achieves a high area under curve of 0.9744 and outperforms the state-of-the-art methods. Promising results are also obtained from experiments on the circ2Disease and circRNADisease databases. Furthermore, the predicted associated circRNAs for hepatocellular carcinoma and gastric cancer are supported by the literature. The code and dataset of this study are available at https://github.com/ABard0/MDGF-MCEC.

Improving segmentation and classification of renal tumors in small sample 3D CT images using transfer learning with convolutional neural networks.

PURPOSE: Computed tomography (CT) images can display internal organs of patients and are particularly suitable for preoperative surgical diagnoses. The increasing demands for computer-aided systems in recent years have facilitated the development of many automated algorithms, especially deep convolutional neural networks, to segment organs and tumors or identify diseases from CT images. However, performances of some systems are highly affected by the amount of training data, while the sizes of medical image data sets, especially three-dimensional (3D) data sets, are usually small. This condition limits the application of deep learning. METHODS: In this study, given a practical clinical data set that has 3D CT images of 20 patients with renal carcinoma, we designed a pipeline employing transfer learning to alleviate the detrimental effect of the small sample size. A dual-channel fine segmentation network (FS-Net) was constructed to segment kidney and tumor regions, with 210 publicly available 3D images from a competition employed during the training phase. We also built discriminative classifiers to classify the benign and malignant tumors based on the segmented regions, where both handcrafted and deep features were tested. RESULTS: Our experimental results showed that the Dice values of segmented kidney and tumor regions were 0.9662 and 0.7685, respectively, which were better than those of state-of-the-art methods. The classification model using radiomics features can classify most of the tumors correctly. CONCLUSIONS: The designed FS-Net was demonstrated to be more effective than simply fine-tuning on the practical small size data set given that the model can borrow knowledge from large auxiliary data without diluting the signal in primary data. For the small data set, radiomics features outperformed deep features in the classification of benign and malignant tumors. This work highlights the importance of architecture design in transfer learning, and the proposed pipeline is anticipated to provide a reference and inspiration for small data analysis.

Interleukin-17A mediates tobacco smoke-induced lung cancer epithelial-mesenchymal transition through transcriptional regulation of ΔNp63α on miR-19.

Interleukin-17A (IL-17A) is an essential inflammatory cytokine in the progress of carcinogenesis. Tobacco smoke (TS) is a major risk factor of lung cancer that influences epithelial-mesenchymal transition (EMT) process. However, the potential mechanism by which IL-17A mediates the progression of lung cancer in TS-induced EMT remains elusive. In the present study, it was revealed that the IL-17A level was elevated in lung cancer tissues, especially in tumor tissues of cases with experience of smoking, and a higher IL-17A level was correlated with induction of EMT in those specimens. Moreover, the expression of ΔNp63α was increased in IL-17A-stimulated lung cancer cells. ΔNp63α functioned as a key oncogene that bound to the miR-17-92 cluster promoter and transcriptionally increased the expression of miR-19 in lung cancer cells. Overexpression of miR-19 promoted EMT in lung cancer with downregulation of E-cadherin and upregulation of N-cadherin, while its inhibition suppressed EMT. Finally, the upregulated levels of IL-17A, ΔNp63α, and miR-19 along with the alteration of EMT-associated biomarkers were found in lung tissues of TS-exposed mice. Taken together, the abovementioned results suggest that IL-17A increases ΔNp63α expression, transcriptionally elevates miR-19 expression, and promotes TS-induced EMT in lung cancer. These findings may provide a new insight for the identification of therapeutic targets for lung cancer.

Cancer-associated fibroblasts enhance the chemoresistance of CD73+ hepatocellular carcinoma cancer cells via HGF-Met-ERK1/2 pathway.

BACKGROUND: Cancer-associated fibroblasts (CAFs) are a major component of hepatocellular carcinoma (HCC) stroma that are critically involved in HCC cancer chemoresistance, but the mechanism has not been elucidated. Previous studies have reported CD73 exerted an immunosuppressive function in cancer. Here, we explored the mechanism by which CAFs regulates CD73+ HCC cells and clarified whether CAFs promote chemoresistance of CD73+ cells. METHODS: We used the co-culture method to study the relationship between CAFs and HCC cells. Immunohistochemistry was applied to evaluate the correlation between α-smooth-muscle actin (α-SMA) and CD73. CD73 mRNA and protein were determined by real-time polymerase chain reaction (RT-PCR) and western blotting, and hepatocyte growth factor (HGF) was assayed by enzyme-linked immunosorbent assay (ELISA). Western blotting was used to explore the regulated pathway of CD73+ HCC. We then knocked down CD73 in cells, and then assessed the effect of CD73 on the apoptosis by flow cytometry. Finally, a sphere formation assay was applied to investigate the stemness of cancer cells, and xenograft tumors in nude mice were built to investigate the tumorigenicity. RESULTS: We found that the proportion of CAFs was positively correlated with CD73 expression in HCC cells. Mechanistically, c-Met and the MEK-ERK1/2 pathway were activated by HGF from CAFs which upregulated CD73 expression in HCC cells. Also, we found that CD73 promote sorafenib and cisplatin resistance in HCC, and CD73+ HCC cells indicated the higher capability of tumorigenicity compared to CD73- HCC cells in vivo. Furthermore, HGF further enhanced the chemoresistant characteristics of CD73+ tumor cells. CONCLUSIONS: Our findings collectively suggest that CD73 is a vital HCC-chemoresistance force controlled by cross-talking between CAFs and HCC cells, thereby establishing CD73 as a potential new therapeutic target for HCC.

SNP rs17079281 decreases lung cancer risk through creating an YY1-binding site to suppress DCBLD1 expression.

Genome-wide association studies (GWAS) have identified numerous genetic variants that are associated with lung cancer risk, but the biological mechanisms underlying these associations remain largely unknown. Here we investigated the functional relevance of a genetic region in 6q22.2 which was identified to be associated with lung cancer risk in our previous GWAS. We performed linkage disequilibrium (LD) analysis and bioinformatic prediction to screen functional SNPs linked to a tagSNP in 6q22.2 loci, followed by two case-control studies and a meta-analysis with 4403 cases and 5336 controls to identify if these functional SNPs were associated with lung cancer risk. A novel SNP rs17079281 in the DCBLD1 promoter was identified to be associated with lung cancer risk in Chinese populations. Compared with those with C allele, patients with T allele had lower risk of adenocarcinoma (adjusted OR = 0.86; 95% CI: 0.80-0.92), but not squamous cell carcinoma (adjusted OR = 0.99; 95% CI: 0.91-1.10), and patients with the C/T or T/T genotype had lower levels of DCBLD1 expression than those with C/C genotype in lung adenocarcinoma tissues. We performed functional assays to characterize its biological relevance. The results showed that the T allele of rs17079281 had higher binding affinity to transcription factor YY1 than the C allele, which suppressed DCBLD1 expression. DCBLD1 behaved like an oncogene, promoting tumor growth by influencing cell cycle progression. These findings suggest that the functional variant rs17079281C>T decreased lung adenocarcinoma risk by creating an YY1-binding site to suppress DCBLD1 expression, which may serve as a biomarker for assessing lung cancer susceptibility.

ImPLoc: a multi-instance deep learning model for the prediction of protein subcellular localization based on immunohistochemistry images.

MOTIVATION: The tissue atlas of the human protein atlas (HPA) houses immunohistochemistry (IHC) images visualizing the protein distribution from the tissue level down to the cell level, which provide an important resource to study human spatial proteome. Especially, the protein subcellular localization patterns revealed by these images are helpful for understanding protein functions, and the differential localization analysis across normal and cancer tissues lead to new cancer biomarkers. However, computational tools for processing images in this database are highly underdeveloped. The recognition of the localization patterns suffers from the variation in image quality and the difficulty in detecting microscopic targets. RESULTS: We propose a deep multi-instance multi-label model, ImPLoc, to predict the subcellular locations from IHC images. In this model, we employ a deep convolutional neural network-based feature extractor to represent image features, and design a multi-head self-attention encoder to aggregate multiple feature vectors for subsequent prediction. We construct a benchmark dataset of 1186 proteins including 7855 images from HPA and 6 subcellular locations. The experimental results show that ImPLoc achieves significant enhancement on the prediction accuracy compared with the current computational methods. We further apply ImPLoc to a test set of 889 proteins with images from both normal and cancer tissues, and obtain 8 differentially localized proteins with a significance level of 0.05. AVAILABILITY AND IMPLEMENTATION: https://github.com/yl2019lw/ImPloc. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Learning complex subcellular distribution patterns of proteins via analysis of immunohistochemistry images.

MOTIVATION: Systematic and comprehensive analysis of protein subcellular location as a critical part of proteomics ('location proteomics') has been studied for many years, but annotating protein subcellular locations and understanding variation of the location patterns across various cell types and states is still challenging. RESULTS: In this work, we used immunohistochemistry images from the Human Protein Atlas as the source of subcellular location information, and built classification models for the complex protein spatial distribution in normal and cancerous tissues. The models can automatically estimate the fractions of protein in different subcellular locations, and can help to quantify the changes of protein distribution from normal to cancer tissues. In addition, we examined the extent to which different annotated protein pathways and complexes showed similarity in the locations of their member proteins, and then predicted new potential proteins for these networks. AVAILABILITY AND IMPLEMENTATION: The dataset and code are available at: www.csbio.sjtu.edu.cn/bioinf/complexsubcellularpatterns. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Traf6 inhibitor boosts antitumor immunity by impeding regulatory T cell migration in Hepa1-6 tumor model.

Tumors escape immune attacks via various mechanisms, among which activation of regulatory pathways in effector immune cells and recruitment of immunosuppressive cells are usually employed. Traf6 is a member of the family of tumor necrosis factor receptor-associated factors and involved in many signaling pathways. While it plays important roles in both tumor biology and immune system, the potential therapeutic role of Traf6 in tumor immunotherapy hasn't ever been assessed. Here, we confirmed the anti-tumor effect of Traf6 inhibitor in Hepa1-6 tumor model. Flow cytometry-based analysis revealed that T cell-mediated antitumor immunity was provoked and the infiltration of Treg cells was restrained when treated with Traf6 inhibitor. Via an in vivo migration assay, we found that Traf6 inhibitor decreased the population of intratumor Tregs by impeding the migration of Tregs towards tumor. Finally, we demonstrated that combination of Traf6 inhibitor and PD-1 blockade could receive a better antitumor efficiency. These results implicated that Traf6 inhibitor could serve as a supplement for immune checkpoint therapy.

Inferring Disease-Associated MicroRNAs Using Semi-supervised Multi-Label Graph Convolutional Networks.

MicroRNAs (miRNAs) play crucial roles in biological processes involved in diseases. The associations between diseases and protein-coding genes (PCGs) have been well investigated, and miRNAs interact with PCGs to trigger them to be functional. We present a computational method, DimiG, to infer miRNA-associated diseases using a semi-supervised Graph Convolutional Network model (GCN). DimiG uses a multi-label framework to integrate PCG-PCG interactions, PCG-miRNA interactions, PCG-disease associations, and tissue expression profiles. DimiG is trained on disease-PCG associations and an interaction network using a GCN, which is further used to score associations between diseases and miRNAs. We evaluate DimiG on a benchmark set from verified disease-miRNA associations. Our results demonstrate that DimiG outperforms the best unsupervised method and is comparable to two supervised methods. Three case studies of prostate cancer, lung cancer, and inflammatory bowel disease further demonstrate the efficacy of DimiG, where top miRNAs predicted by DimiG are supported by literature.

The inherited variations of a p53-responsive enhancer in 13q12.12 confer lung cancer risk by attenuating TNFRSF19 expression.

BACKGROUND: Inherited factors contribute to lung cancer risk, but the mechanism is not well understood. Defining the biological consequence of GWAS hits in cancers is a promising strategy to elucidate the inherited mechanisms of cancers. The tag-SNP rs753955 (A>G) in 13q12.12 is highly associated with lung cancer risk in the Chinese population. Here, we systematically investigate the biological significance and the underlying mechanism behind 13q12.12 risk locus in vitro and in vivo. RESULTS: We characterize a novel p53-responsive enhancer with lung tissue cell specificity in a 49-kb high linkage disequilibrium block of rs753955. This enhancer harbors 3 highly linked common inherited variations (rs17336602, rs4770489, and rs34354770) and six p53 binding sequences either close to or located between the variations. The enhancer effectively protects normal lung cell lines against pulmonary carcinogen NNK-induced DNA damages and malignant transformation by upregulating TNFRSF19 through chromatin looping. These variations significantly weaken the enhancer activity by affecting its p53 response, especially when cells are exposed to NNK. The effect of the mutant enhancer alleles on TNFRSF19 target gene in vivo is supported by expression quantitative trait loci analysis of 117 Chinese NSCLC samples and GTEx data. Differentiated expression of TNFRSF19 and its statistical significant correlation with tumor TNM staging and patient survival indicate a suppressor role of TNFRSF19 in lung cancer. CONCLUSION: This study provides evidence of how the inherited variations in 13q12.12 contribute to lung cancer risk, highlighting the protective roles of the p53-responsive enhancer-mediated TNFRSF19 activation in lung cells under carcinogen stress.

TRAF6 directs FOXP3 localization and facilitates regulatory T-cell function through K63-linked ubiquitination.

Regulatory T cells (Tregs) are crucial mediators of immune control. The characteristic gene expression and suppressive functions of Tregs depend considerably on the stable expression and activity of the transcription factor FOXP3. Transcriptional regulation of the Foxp3 gene has been studied in depth, but both the expression and function of this factor are also modulated at the protein level. However, the molecular players involved in posttranslational FOXP3 regulation are just beginning to be elucidated. Here, we found that TRAF6-deficient Tregs were dysfunctional in vivo; mice with Treg-restricted deletion of TRAF6 were resistant to implanted tumors and displayed enhanced anti-tumor immunity. We further determined that FOXP3 undergoes K63-linked ubiquitination at lysine 262 mediated by the E3 ligase TRAF6. In the absence of TRAF6 activity or upon mutation of the ubiquitination site, FOXP3 displayed aberrant, perinuclear accumulation and disrupted regulatory function. Thus, K63-linked ubiquitination by TRAF6 ensures proper localization of FOXP3 and facilitates the transcription factor's gene-regulating activity in Tregs. These results implicate TRAF6 as a key posttranslational, Treg-stabilizing regulator that may be targeted in novel tolerance-breaking therapies.

LabCaS for Ranking Potential Calpain Substrate Cleavage Sites from Amino Acid Sequence.

Calpains are a family of Ca2+-dependent cysteine proteases involved in many important biological processes, where they selectively cleave relevant substrates at specific cleavage sites to regulate the function of the substrate proteins. Presently, our knowledge about the function of calpains and the mechanism of substrate cleavage is still limited due to the fact that the experimental determination and validation on calpain bindings are usually laborious and expensive. This chapter describes LabCaS, an algorithm that is designed for predicting the calpain substrate cleavage sites from amino acid sequences. LabCaS is built on a conditional random field (CRF) statistic model, which trains the cleavage site prediction on multiple features of amino acid residue preference, solvent accessibility information, pair-wise alignment similarity score, secondary structure propensity, and physical-chemistry properties. Large-scale benchmark tests have shown that LabCaS can achieve a reliable recognition of the cleavage sites for most calpain proteins with an average AUC score of 0.862. Due to the fast speed and convenience of use, the protocol should find its usefulness in large-scale calpain-based function annotations of the newly sequenced proteins. The online web server of LabCaS is freely available at http://www.csbio.sjtu.edu.cn/bioinf/LabCaS .

An Exhaustive Search Algorithm to Aid NMR-Based Structure Determination of Rotationally Symmetric Transmembrane Oligomers.

Nuclear magnetic resonance (NMR) has been an important source of structural restraints for solving structures of oligomeric transmembrane domains (TMDs) of cell surface receptors and viral membrane proteins. In NMR studies, oligomers are assembled using inter-protomer distance restraints. But, for oligomers that are higher than dimer, these distance restraints all have two-fold directional ambiguity, and resolving such ambiguity often requires time-consuming trial-and-error calculations using restrained molecular dynamics (MD) with simulated annealing (SA). We report an Exhaustive Search algorithm for Symmetric Oligomer (ExSSO), which can perform near-complete search of the symmetric conformational space in a very short time. In this approach, the predetermined protomer model is subject to full angular and spatial search within the symmetry space. This approach, which can be applied to any rotationally symmetric oligomers, was validated using the structures of the Fas death receptor, the HIV-1 gp41 fusion protein, the influenza proton channel, and the MCU pore. The algorithm is able to generate approximate oligomer solutions quickly as initial inputs for further refinement using the MD/SA method.

Dynamic regulation of CD28 conformation and signaling by charged lipids and ions.

CD28 provides an essential costimulatory signal for T cell activation, and its function is critical in antitumor immunity. However, the molecular mechanism of CD28 transmembrane signaling remains elusive. Here we show that the conformation and signaling of CD28 are regulated by two counteractive charged factors, acidic phospholipids and Ca2+ ions. NMR spectroscopy analyses showed that acidic phospholipids can sequester CD28 signaling motifs within the membrane, thereby limiting CD28 basal signaling. T cell receptor (TCR) activation induced an increase in the local Ca2+ concentration around CD28, and Ca2+ directly disrupted CD28-lipid interaction, leading to opening and signaling of CD28. We observed that the TCR, Ca2+, and CD28 together form a dual-positive-feedback circuit that substantially amplifies T cell signaling and thus increases antigen sensitivity. This work unravels a new regulatory mechanism for CD28 signaling and thus contributes to the understanding of the dependence of costimulation signaling on TCR signaling and the high sensitivity of T cells.

TargetM6A: Identifying N6-Methyladenosine Sites From RNA Sequences via Position-Specific Nucleotide Propensities and a Support Vector Machine.

As one of the most ubiquitous post-transcriptional modifications of RNA, N6-methyladenosine ( [Formula: see text]) plays an essential role in many vital biological processes. The identification of [Formula: see text] sites in RNAs is significantly important for both basic biomedical research and practical drug development. In this study, we designed a computational-based method, called TargetM6A, to rapidly and accurately target [Formula: see text] sites solely from the primary RNA sequences. Two new features, i.e., position-specific nucleotide/dinucleotide propensities (PSNP/PSDP), are introduced and combined with the traditional nucleotide composition (NC) feature to formulate RNA sequences. The extracted features are further optimized to obtain a much more compact and discriminative feature subset by applying an incremental feature selection (IFS) procedure. Based on the optimized feature subset, we trained TargetM6A on the training dataset with a support vector machine (SVM) as the prediction engine. We compared the proposed TargetM6A method with existing methods for predicting [Formula: see text] sites by performing stringent jackknife tests and independent validation tests on benchmark datasets. The experimental results show that the proposed TargetM6A method outperformed the existing methods for predicting [Formula: see text] sites and remarkably improved the prediction performances, with MCC = 0.526 and AUC = 0.818. We also provided a user-friendly web server for TargetM6A, which is publicly accessible for academic use at http://csbio.njust.edu.cn/bioinf/TargetM6A.

Improving N(6)-methyladenosine site prediction with heuristic selection of nucleotide physical-chemical properties.

N(6)-methyladenosine (m(6)A) is one of the most common and abundant post-transcriptional RNA modifications found in viruses and most eukaryotes. m(6)A plays an essential role in many vital biological processes to regulate gene expression. Because of its widespread distribution across the genomes, the identification of m(6)A sites from RNA sequences is of significant importance for better understanding the regulatory mechanism of m(6)A. Although progress has been achieved in m(6)A site prediction, challenges remain. This article aims to further improve the performance of m(6)A site prediction by introducing a new heuristic nucleotide physical-chemical property selection (HPCS) algorithm. The proposed HPCS algorithm can effectively extract an optimized subset of nucleotide physical-chemical properties under the prescribed feature representation for encoding an RNA sequence into a feature vector. We demonstrate the efficacy of the proposed HPCS algorithm under different feature representations, including pseudo dinucleotide composition (PseDNC), auto-covariance (AC), and cross-covariance (CC). Based on the proposed HPCS algorithm, we implemented an m(6)A site predictor, called M6A-HPCS, which is freely available at http://csbio.njust.edu.cn/bioinf/M6A-HPCS. Experimental results over rigorous jackknife tests on benchmark datasets demonstrated that the proposed M6A-HPCS achieves higher success rates and outperforms existing state-of-the-art sequence-based m(6)A site predictors.

Genetic variants within microRNA-binding site of RAD51B are associated with risk of cervical cancer in Chinese women.

RAD51B plays a central role in homologous recombinational repair (HRR) of DNA double-strand breaks (DSBs), which is important to prevent genomic instability, a hallmark of cancer. Recent studies suggested that common genetic variants of RAD51B may contribute to cancer susceptibility. In this study, we aimed to investigate whether potentially functional variants within miRNA-binding sites of RAD51B are associated with risk of cervical cancer. A total of 1486 cervical cancer patients and 1536 cancer-free controls were enrolled, and two genetic variants, rs963917 (A > G) and rs963918 (T > C), were genotyped in all participants. Using multivariate logistic regression analyses, we found that G allele of rs963917 conferred lower risk of cervical cancer compared to A allele (adjusted OR = 0.89, 95% CI = 0.80-0.99, P = 0.039). Similarly, rs963918 allele C was associated with a decreased risk for cervical cancer compared with allele T (adjusted OR = 0.84, 95% CI = 0.74-0.94, P = 0.004). Haplotype analyses showed that haplotype GC was also correlated with lower risk (OR = 0.83, 95% CI = 0.73-0.95, P = 0.005) compared with the most common haplotype AT. In summary, our study suggested that miRNA-binding site genetic variants of RAD51B may modify the susceptibility to cervical cancer, which is important to identify individuals with differential risk for this malignancy and to improve the effectiveness of preventive intervention.

Accurate disulfide-bonding network predictions improve ab initio structure prediction of cysteine-rich proteins.

MOTIVATION: Cysteine-rich proteins cover many important families in nature but there are currently no methods specifically designed for modeling the structure of these proteins. The accuracy of disulfide connectivity pattern prediction, particularly for the proteins of higher-order connections, e.g., >3 bonds, is too low to effectively assist structure assembly simulations. RESULTS: We propose a new hierarchical order reduction protocol called Cyscon for disulfide-bonding prediction. The most confident disulfide bonds are first identified and bonding prediction is then focused on the remaining cysteine residues based on SVR training. Compared with purely machine learning-based approaches, Cyscon improved the average accuracy of connectivity pattern prediction by 21.9%. For proteins with more than 5 disulfide bonds, Cyscon improved the accuracy by 585% on the benchmark set of PDBCYS. When applied to 158 non-redundant cysteine-rich proteins, Cyscon predictions helped increase (or decrease) the TM-score (or RMSD) of the ab initio QUARK modeling by 12.1% (or 14.4%). This result demonstrates a new avenue to improve the ab initio structure modeling for cysteine-rich proteins. AVAILABILITY AND IMPLEMENTATION: http://www.csbio.sjtu.edu.cn/bioinf/Cyscon/ CONTACT: zhng@umich.edu or hbshen@sjtu.edu.cn. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.