A novel missense-mutation-related feature extraction scheme for 'driver' mutation identification

Radiogenomics-Based Risk Prediction of Glioblastoma Multiforme with Clinical Relevance

Glioblastoma multiforme (GBM)is the most common and aggressive primary brain tumor. Although temozolomide (TMZ)-based radiochemotherapy improves overall GBM patients' survival, it also increases the frequency of false positive post-treatment magnetic resonance imaging (MRI) assessments for tumor progression. Pseudo-progression (PsP) is a treatment-related reaction with an increased contrast-enhancing lesion size at the tumor site or resection margins miming tumor recurrence on MRI. The accurate and reliable prognostication of GBM progression is urgently needed in the clinical management of GBM patients. Clinical data analysis indicates that the patients with PsP had superior overall and progression-free survival rates. In this study, we aimed to develop a prognostic model to evaluate the tumor progression potential of GBM patients following standard therapies. We applied a dictionary learning scheme to obtain imaging features of GBM patients with PsP or true tumor progression (TTP) from the Wake dataset. Based on these radiographic features, we conducted a radiogenomics analysis to identify the significantly associated genes. These significantly associated genes were used as features to construct a 2YS (2-year survival rate) logistic regression model. GBM patients were classified into low- and high-survival risk groups based on the individual 2YS scores derived from this model. We tested our model using an independent The Cancer Genome Atlas Program (TCGA) dataset and found that 2YS scores were significantly associated with the patient's overall survival. We used two cohorts of the TCGA data to train and test our model. Our results show that the 2YS scores-based classification results from the training and testing TCGA datasets were significantly associated with the overall survival of patients. We also analyzed the survival prediction ability of other clinical factors (gender, age, KPS (Karnofsky performance status), normal cell ratio) and found that these factors were unrelated or weakly correlated with patients' survival. Overall, our studies have demonstrated the effectiveness and robustness of the 2YS model in predicting the clinical outcomes of GBM patients after standard therapies.

Differential diagnosis of pancreatic cystic neoplasms through a radiomics-assisted system

Pancreatic cystic neoplasms (PCNs) are a group of heterogeneous diseases with distinct prognosis. Existing differential diagnosis methods require invasive biopsy or prolonged monitoring. We sought to develop an inexpensive, non-invasive differential diagnosis system for PCNs based on radiomics features and clinical characteristics for a higher total PCN screening rate. We retrospectively analyzed computed tomography images and clinical data from 129 patients with PCN, including 47 patients with intraductal papillary mucinous neoplasms (IPMNs), 49 patients with serous cystadenomas (SCNs), and 33 patients with mucinous cystic neoplasms (MCNs). Six clinical characteristics and 944 radiomics features were tested, and nine features were finally selected for model construction using DXScore algorithm. A five-fold cross-validation algorithm and a test group were applied to verify the results. In the five-fold cross-validation section, the AUC value of our model was 0.8687, and the total accuracy rate was 74.23%, wherein the accuracy rates of IPMNs, SCNs, and MCNs were 74.26%, 78.37%, and 68.00%, respectively. In the test group, the AUC value was 0.8462 and the total accuracy rate was 73.61%. In conclusion, our research constructed an end-to-end powerful PCN differential diagnosis system based on radiomics method, which could assist decision-making in clinical practice.

Machine learning methods for prediction of cancer driver genes: a survey paper

Identifying the genes and mutations that drive the emergence of tumors is a critical step to improving our understanding of cancer and identifying new directions for disease diagnosis and treatment. Despite the large volume of genomics data, the precise detection of driver mutations and their carrying genes, known as cancer driver genes, from the millions of possible somatic mutations remains a challenge. Computational methods play an increasingly important role in discovering genomic patterns associated with cancer drivers and developing predictive models to identify these elements. Machine learning (ML), including deep learning, has been the engine behind many of these efforts and provides excellent opportunities for tackling remaining gaps in the field. Thus, this survey aims to perform a comprehensive analysis of ML-based computational approaches to identify cancer driver mutations and genes, providing an integrated, panoramic view of the broad data and algorithmic landscape within this scientific problem. We discuss how the interactions among data types and ML algorithms have been explored in previous solutions and outline current analytical limitations that deserve further attention from the scientific community. We hope that by helping readers become more familiar with significant developments in the field brought by ML, we may inspire new researchers to address open problems and advance our knowledge towards cancer driver discovery.

Differentiating TP53 Mutation Status in Pancreatic Ductal Adenocarcinoma Using Multiparametric MRI-Derived Radiomics

Objectives

This study assessed the preoperative prediction of TP53 status based on multiparametric magnetic resonance imaging (mpMRI) radiomics extracted from two-dimensional (2D) and 3D images.

Methods

57 patients with pancreatic cancer who underwent preoperative MRI were included. The diagnosis and TP53 gene test were based on resections. Of the 57 patients included 37 mutated TP53 genes and the remaining 20 had wild-type TP53 genes. Two radiologists performed manual tumour segmentation on seven different MRI image acquisition sequences per patient, including multi-phase [pre-contrast, late arterial phase (ap), portal venous phase, and delayed phase] dynamic contrast enhanced (DCE) T1-weighted imaging, T2-weighted imaging (T2WI), Diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC). PyRadiomics-package was used to generate 558 two-dimensional (2D) and 994 three-dimensional (3D) image features. Models were constructed by support vector machine (SVM) for differentiating TP53 status and DX score method were used for feature selection. The evaluation of the model performance included area under the curve (AUC), accuracy, calibration curves, and decision curve analysis.

Results

The 3D ADC-ap-DWI-T2WI model with 11 selected features yielded the best performance for differentiating TP53 status, with accuracy = 0.91 and AUC = 0.96. The model showed the good calibration. The decision curve analysis indicated that the radiomics model had clinical utility.

Conclusions

A non-invasive and quantitative mpMRI-based radiomics model can accurately predict TP53 mutation status in pancreatic cancer patients and contribute to the precision treatment.

Inferring subgroup-specific driver genes from heterogeneous cancer samples via subspace learning with subgroup indication

MOTIVATION

Detecting driver genes from gene mutation data is a fundamental task for tumorigenesis research. Due to the fact that cancer is a heterogeneous disease with various subgroups, subgroup-specific driver genes are the key factors in the development of precision medicine for heterogeneous cancer. However, the existing driver gene detection methods are not designed to identify subgroup specificities of their detected driver genes, and therefore cannot indicate which group of patients is associated with the detected driver genes, which is difficult to provide specifically clinical guidance for individual patients.

RESULTS

By incorporating the subspace learning framework, we propose a novel bioinformatics method called DriverSub, which can efficiently predict subgroup-specific driver genes in the situation where the subgroup annotations are not available. When evaluated by simulation datasets with known ground truth and compared with existing methods, DriverSub yields the best prediction of driver genes and the inference of their related subgroups. When we apply DriverSub on the mutation data of real heterogeneous cancers, we can observe that the predicted results of DriverSub are highly enriched for experimentally validated known driver genes. Moreover, the subgroups inferred by DriverSub are significantly associated with the annotated molecular subgroups, indicating its capability of predicting subgroup-specific driver genes.

AVAILABILITY AND IMPLEMENTATION

The source code is publicly available at https://github.com/JianingXi/DriverSub.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

miRactDB characterizes miRNA-gene relation switch between normal and cancer tissues across pan-cancer

It has been increasingly accepted that microRNA (miRNA) can both activate and suppress gene expression, directly or indirectly, under particular circumstances. Yet, a systematic study on the switch in their interaction pattern between activation and suppression and between normal and cancer conditions based on multi-omics evidences is not available. We built miRactDB, a database for miRNA-gene interaction, at https://ccsm.uth.edu/miRactDB, to provide a versatile resource and platform for annotation and interpretation of miRNA-gene relations. We conducted a comprehensive investigation on miRNA-gene interactions and their biological implications across tissue types in both tumour and normal conditions, based on TCGA, CCLE and GTEx databases. We particularly explored the genetic and epigenetic mechanisms potentially contributing to the positive correlation, including identification of miRNA binding sites in the gene coding sequence (CDS) and promoter regions of partner genes. Integrative analysis based on this resource revealed that top-ranked genes derived from TCGA tumour and adjacent normal samples share an overwhelming part of biological processes, which are quite different than those from CCLE and GTEx. The most active miRNAs predicted to target CDS and promoter regions are largely overlapped. These findings corroborate that adjacent normal tissues might have undergone significant molecular transformations towards oncogenesis before phenotypic and histological change; and there probably exists a small yet critical set of miRNAs that profoundly influence various cancer hallmark processes. miRactDB provides a unique resource for the cancer and genomics communities to screen, prioritize and rationalize their candidates of miRNA-gene interactions, in both normal and cancer scenarios.

A novel microRNA signature for pathological grading in lung adenocarcinoma based on TCGA and GEO data

Lung adenocarcinoma (LUAD) is one of the most common types of lung cancer and its poor prognosis largely depends on the tumor pathological stage. Critical roles of microRNAs (miRNAs) have been reported in the tumorigenesis and progression of lung cancer. However, whether the differential expression pattern of miRNAs could be used to distinguish early-stage (stage I) from mid-late-stage (stages II–IV) LUAD tumors is still unclear. In this study, clinical information and miRNA expression profiles of patients with LUAD were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus databases. TCGA-LUAD (n=470) dataset was used for model training and validation, and the GSE62182 (n=94) and GSE83527 (n=36) datasets were used as external independent test datasets. The diagnostic model was created through miRNA feature selection followed by SVM classifier and was confirmed by 5-fold cross-validation. A receiver operating characteristic curve was calculated to evaluate the accuracy and robustness of the model. Using the DX score and LIBSVM tool, a 16-miRNA signature that could distinguish LUAD pathological stages was identified. The area under the curve rates were 0.62 [95% confidence interval (CI): 0.56–0.67], 0.66 (95% CI: 0.54–0.76) and 0.63 (95% CI: 0.43–0.82) in TCGA-LUAD internal validation dataset, the GSE62182 external validation dataset, and the GSE83527 external validation dataset, respectively. Kyoto Encyclopedia of Genes and Genomes and Gene Ontology enrichment analyses suggested that the target genes of the 16-miRNA signature were mainly involved in metabolic pathways. The present findings demonstrate that a 16-miRNA signature could serve as a promising diagnostic biomarker for pathological staging in LUAD.

Association of miR-146a, miR-149 and miR-196a2 polymorphisms with neuroblastoma risk in Eastern Chinese population: a three-center case-control study

Neuroblastoma is one of the most common malignancy in childhood, which originates from the developing sympathetic nervous system. Single nucleotide polymorphisms (SNPs) in primary miRNA (pri-miRNA) have shown to associate with cancer susceptibility, including neuroblastoma. Three precursor miRNA (pre-miRNA) SNPs (pre-miR-146a rs2910164, pre-miR-149 rs2292832 and pre-miR-196a2 rs11614913) were found to contribute to pathogenesis of various diseases. Here, to evaluate the association among these three pre-miRNA SNPs and neuroblastoma susceptibility in Eastern Chinese children, we carried out a three-center case-control study involving 312 neuroblastoma cases and 762 healthy controls. Odds ratios (ORs) and 95% confidence intervals (CIs) were used to assess the association of these three polymorphisms with neuroblastoma risk. However, no significant association was observed among these three SNPs and neuroblastoma susceptibility, in either overall or subgroups analysis by tumor sites, gender and age. Further larger studies consisting of diverse ethnic populations are required to clarify the associations among these three pre-miRNAs polymorphisms and neuroblastoma risk.

Pan-cancer analysis on microRNA-associated gene activation

Background

While microRNAs (miRNAs) were widely considered to repress target genes at mRNA and/or protein levels, emerging evidence from in vitro experiments has shown that miRNAs can also activate gene expression in particular contexts. However, this counterintuitive observation has rarely been reported or interpreted in in vivo conditions.

Methods

We systematically explored the positive correlation between miRNA and gene expressions and its potential implications in tumorigenesis, based on 8375 patient samples across 31 major human cancers from The Cancer Genome Atlas (TCGA).

Findings

We found that positive miRNA-gene correlations are surprisingly prevalent and consistent across cancer types, and show distinct patterns than negative correlations. The top-ranked positive correlations are significantly involved in the immune cell differentiation and cell membrane signaling related processes, and display strong power in stratifying patients in terms of survival rate. Although intragenic miRNAs generally tend to co-express with their host genes, a substantial portion of miRNAs shows no obvious correlation with their host gene plausibly due to non-conservation. A miRNA can upregulate a gene by inhibiting its upstream suppressor, or shares transcription factors with that gene, both leading to positive correlation. The miRNA/gene sites associated with the top-ranked positive correlations are more likely to form super-enhancers compared to randomly chosen pairs. Wet-lab experiments revealed that positive correlations partially remain in in vitro condition.

Interpretation

Our study brings new insights into the critical role of miRNA in gene regulation and the complex mechanisms underlying miRNA functions, and reveals both biological and clinical significance of miRNA-associated gene activation.

Challenges in the introduction of next-generation sequencing (NGS) for diagnostics of myeloid malignancies into clinical routine use

Given the vast phenotypic and genetic heterogeneity of acute and chronic myeloid malignancies, hematologists have eagerly awaited the introduction of next-generation sequencing (NGS) into the routine diagnostic armamentarium to enable a more differentiated disease classification, risk stratification, and improved therapeutic decisions. At present, an increasing number of hematologic laboratories are in the process of integrating NGS procedures into the diagnostic algorithms of patients with acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and myeloproliferative neoplasms (MPNs). Inevitably accompanying such developments, physicians and molecular biologists are facing unexpected challenges regarding the interpretation and implementation of molecular genetic results derived from NGS in myeloid malignancies. This article summarizes typical challenges that may arise in the context of NGS-based analyses at diagnosis and during follow-up of myeloid malignancies.

Detection of Combinatorial Mutational Patterns in Human Cancer Genomes by Exclusivity Analysis

Cancer genes may tend to mutate in a co-mutational or mutually exclusive manner in a tumor sample of a specific cancer, which constitute two known combinatorial mutational patterns for a given gene set. Previous studies have established that genes functioning in different signaling pathways can mutate in the same sample, i.e., a tumor from one patient, while genes operating in the same pathway are rarely mutated in the same cancer genome. Therefore, reliable identification of combinatorial mutational patterns of candidate cancer genes has important ramifications in inferring signaling network modules in a particular cancer type. While algorithms for discovering mutated driver pathways based on mutual exclusivity of mutations in cancer genes have been proposed, a systematic pipeline for identifying both co-mutational and mutually exclusive patterns with rational significance estimation is still lacking. Here, we describe a reliable framework with detailed procedures to simultaneously explore both combinatorial mutational patterns from public cross-sectional gene mutation data.

Intraglomerular crosstalk elaborately regulates podocyte injury and repair in diabetic patients: insights from a 3D multiscale modeling study

Podocytes are mainly involved in the regulation of glomerular filtration rate (GFR) under physiological condition. Podocyte depletion is a crucial pathological alteration in diabetic nephropathy (DN) and results in a broad spectrum of clinical syndromes such as protein urine and renal insufficiency. Recent studies indicate that depleted podocytes can be regenerated via differentiation of the parietal epithelial cells (PECs), which serve as the local progenitors of podocytes. However, the podocyte regeneration process is regulated by a complicated mechanism of cell-cell interactions and cytokine stimulations, which has been studied in a piecemeal manner rather than systematically. To address this gap, we developed a high-resolution multi-scale multi-agent mathematical model in 3D, mimicking the in situ glomerulus anatomical structure and micro-environment, to simulate the podocyte regeneration process under various cytokine perturbations in healthy and diabetic conditions. Our model showed that, treatment with pigment epithelium derived factor (PEDF) or insulin-like growth factor-1 (IGF-1) alone merely ameliorated the glomerulus injury, while co-treatment with both cytokines replenished the damaged podocyte population gradually. In addition, our model suggested that continuous administration of PEDF instead of a bolus injection sustained the regeneration process of podocytes. Part of the results has been validated by our in vivo experiments. These results indicated that amelioration of the glomerular stress by PEDF and promotion of PEC differentiation by IGF-1 are equivalently critical for podocyte regeneration. Our 3D multi-scale model represents a powerful tool for understanding the signaling regulation and guiding the design of cytokine therapies in promoting podocyte regeneration.

Identification of biomarkers for pseudo and true progression of GBM based on radiogenomics study

The diagnosis for pseudoprogression (PsP) and true tumor progression (TTP) of GBM is a challenging task in clinical practices. The purpose of this study is to identify potential genetic biomarkers associated with PsP and TTP based on the clinical records, longitudinal imaging features, and genomics data. We are the first to introduce the radiogenomics approach to identify candidate genes for PsP and TTP of GBM. Specifically, a novel longitudinal sparse regression model was developed to construct the relationship between gene expression and imaging features. The imaging features were extracted from tumors along the longitudinal MRI and provided diagnostic information of PsP and TTP. The 33 candidate genes were selected based on their association with the imaging features, reflecting their relation with the development of PsP and TTP. We then conducted biological relevance analysis for 33 candidate genes to identify the potential biomarkers, i.e., Interferon regulatory factor (IRF9) and X-ray repair cross-complementing gene (XRCC1), which were involved in the cancer suppression and prevention, respectively. The IRF9 and XRCC1 were further independently validated in the TCGA data. Our results provided the first substantial evidence that IRF9 and XRCC1 can serve as the potential biomarkers for the development of PsP and TTP.

Association between NEFL Gene Polymorphisms and Neuroblastoma Risk in Chinese Children: A Two-Center Case-Control Study

Neuroblastoma is a lethal tumor that mainly occurs in children. To date, the genetic etiology of sporadic neuroblastoma remains obscure. A previous study identified three neuroblastoma susceptibility loci (rs11994014 G>A, rs2979704 T>C, rs1059111 A>T) in neurofilament light (NEFL) gene. Here, we attempted to evaluate the contributions of these three single nucleotide polymorphisms to neuroblastoma susceptibility in Chinese children. We genotyped these three polymorphisms using subjects from Guangdong province (256 cases and 531 controls) and Henan province (118 cases and 281 controls). Logistic regression models were performed to generate odds ratios and 95% confidence intervals to access the association of these three polymorphisms with neuroblastoma risk. Overall, we failed to provide any evidence supporting the association between these three polymorphisms and neuroblastoma susceptibility, either in single center population or in the combined population. Moreover, such null association was also observed when the samples were stratified by age, gender, tumor sites, and clinical stages. In the future, larger samples from different ethnicities are needed to clarify the role of NEFL gene polymorphisms in neuroblastoma risk.

Applications of Support Vector Machine (SVM) Learning in Cancer Genomics

Machine learning with maximization (support) of separating margin (vector), called support vector machine (SVM) learning, is a powerful classification tool that has been used for cancer genomic classification or subtyping. Today, as advancements in high-throughput technologies lead to production of large amounts of genomic and epigenomic data, the classification feature of SVMs is expanding its use in cancer genomics, leading to the discovery of new biomarkers, new drug targets, and a better understanding of cancer driver genes. Herein we reviewed the recent progress of SVMs in cancer genomic studies. We intend to comprehend the strength of the SVM learning and its future perspective in cancer genomic applications.

Drug repurposing: An approach to tackle drug resistance in S. typhimurium

Drug resistant S. typhimurium pose important public health problem. The development of effective drugs with novel mechanism(s) of action is needed to overcome issues pertaining to drug resistance. Drug repurposing based on computational analyses is considered a viable alternative strategy to circumvent this issue. In this context, 1309 FDA-approved drugs molecules from Mantra 2.0 database were analyzed for this study, against S. typhimurium. Sixteen compounds having similar profiles of gene expression as quinolones were identified from the database, Mantra 2.0. Further, the pharmacophore characteristics of each resultant molecule were identified and compared with the features of nalidixic acid, using the PharamGist program. Subsequently, the activities of these compounds against S. typhimurium DNA gyrase were identified, using molecular docking study. Side effects analysis was also performed for the identified compounds. Molecular dynamics simulation was carried out for the compound to validate its binding efficiency. Further, characterization of screened compound revealed IC50 values in micromolar concentration range, of which flufenamic acid showed comparable in vitro activity alongside ciprofloxacin and nalidixic acid. Thus represent interesting starting points for further optimization against S. typhimurium infections. It may be noted that the results we have obtained are the first experimental evidence of flufenamic acid activity against S. typhimurium.

Significant association of the EXO1 rs851797 polymorphism with clinical outcome of ovarian cancer

BACKGROUND

Exonuclease 1 (EXO1), one of DNA mismatch repair pathway genes, functions in maintaining genomic stability and affects tumor progression. We hypothesized that genetic variations in EXO1 may predict clinical outcomes in epithelial ovarian cancer (EOC).

METHODS

In this cohort study with 1,030 consecutive EOC patients, we genotyped four potentially functional polymorphisms in EXO1 by the Taqman assay and evaluated their associations with patients' survival.

RESULTS

Using multivariate Cox proportional hazards regression models, we found that rs851797AG/GG genotypes were significantly associated with recurrence and cancer death (HR =1.30 and 1.38, 95% CI =1.11-1.52 and 1.02-1.88, respectively). Kaplan-Meier survival estimates showed that patients who carried rs851797AG/GG genotypes had poorer progression-free survival and poorer overall survival, compared with rs851797AA genotype carriers (log-rank test, P=0.002 and 0.025, respectively). Moreover, patients with older age at menophania, advanced stage tumor, or being received incomplete cytoreduction were more likely to be recurrent and dead.

CONCLUSION

EXO1 rs851797 polymorphism can predict the clinical outcomes in EOC patients. In addition, age at menophania, FIGO stage, and complete cytoreduction might be independently prognostic factors of ovarian cancer. Large studies with functional experiments are warranted to validate these findings.

Stratification of pseudoprogression and true progression of glioblastoma multiform based on longitudinal diffusion tensor imaging without segmentation

PURPOSE

Pseudoprogression (PsP) can mimic true tumor progression (TTP) on magnetic resonance imaging in patients with glioblastoma multiform (GBM). The phenotypical similarity between PsP and TTP makes it a challenging task for physicians to distinguish these entities. So far, no approved biomarkers or computer-aided diagnosis systems have been used clinically for this purpose.

METHODS

To address this challenge, the authors developed an objective classification system for PsP and TTP based on longitudinal diffusion tensor imaging. A novel spatio-temporal discriminative dictionary learning scheme was proposed to differentiate PsP and TTP, thereby avoiding segmentation of the region of interest. The authors constructed a novel discriminative sparse matrix with the classification-oriented dictionary learning approach by excluding the shared features of two categories, so that the pooled features captured the subtle difference between PsP and TTP. The most discriminating features were then identified from the pooled features by their feature scoring system. Finally, the authors stratified patients with GBM into PsP and TTP by a support vector machine approach. Tenfold cross-validation (CV) and the area under the receiver operating characteristic (AUC) were used to assess the robustness of the developed system.

RESULTS

The average accuracy and AUC values after ten rounds of tenfold CV were 0.867 and 0.92, respectively. The authors also assessed the effects of different methods and factors (such as data types, pooling techniques, and dimensionality reduction approaches) on the performance of their classification system which obtained the best performance.

CONCLUSIONS

The proposed objective classification system without segmentation achieved a desirable and reliable performance in differentiating PsP from TTP. Thus, the developed approach is expected to advance the clinical research and diagnosis of PsP and TTP.

The TP53 gene rs1042522 C>G polymorphism and neuroblastoma risk in Chinese children

TP53, a tumor suppressor gene, plays a critical role in cell cycle control, apoptosis, and DNA damage repair. Previous studies have indicated that the TP53 gene Arg72Pro (rs1042522 C>G) polymorphism is associated with susceptibility to various types of cancer. We evaluated the association of the TP53 gene rs1042522 C>G polymorphism with neuroblastoma susceptibility in a hospital-based study among the Chinese Han population. Enrolled were 256 patients and 531 controls. Odds ratios (ORs) and 95% confidence intervals (CIs) generated using logistic regression models were used to determine the strength of the association of interest. No association was detected between rs1042522 C>G polymorphism and neuroblastoma risk. In our stratification analysis of age, gender, sites of origin, and clinical stages, we observed that subjects with rs1042522 CG/GG genotypes had a lower risk of developing neuroblastoma in the mediastinum (Adjusted OR=0.52, 95% CI=0.33-0.82, P=0.005) than those carrying the CC genotype. These results indicate that TP53 gene rs1042522 C>G polymorphism may exert a weak and site-specific effect on neuroblastoma risk in Southern Chinese children and warrant further confirmation.

A Systematic Approach to Predicting Spring Force for Sagittal Craniosynostosis Surgery

Spring-assisted surgery (SAS) can effectively treat scaphocephaly by reshaping crania with the appropriate spring force. However, it is difficult to accurately estimate spring force without considering biomechanical properties of tissues. This study presents and validates a reliable system to accurately predict the spring force for sagittal craniosynostosis surgery. The authors randomly chose 23 patients who underwent SAS and had been followed for at least 2 years. An elastic model was designed to characterize the biomechanical behavior of calvarial bone tissue for each individual. After simulating the contact force on accurate position of the skull strip with the springs, the finite element method was applied to calculating the stress of each tissue node based on the elastic model. A support vector regression approach was then used to model the relationships between biomechanical properties generated from spring force, bone thickness, and the change of cephalic index after surgery. Therefore, for a new patient, the optimal spring force can be predicted based on the learned model with virtual spring simulation and dynamic programming approach prior to SAS. Leave-one-out cross-validation was implemented to assess the accuracy of our prediction. As a result, the mean prediction accuracy of this model was 93.35%, demonstrating the great potential of this model as a useful adjunct for preoperative planning tool.

MDM4 rs4245739 A > C polymorphism correlates with reduced overall cancer risk in a meta-analysis of 69477 subjects

Mouse double minute 4 (MDM4) is a p53-interacting oncoprotein that plays an important role in the p53 tumor suppressor pathway. The common rs4245739 A > C polymorphism creates a miR-191 binding site in the MDM4 gene transcript. Numerous studies have investigated the association between this MDM4 polymorphism and cancer risk, but have failed to reach a definitive conclusion. To address this issue, we conducted a meta-analysis by selecting eligible studies from MEDLINE, EMBASE, and Chinese Biomedical databases. Odds ratios (ORs) and 95% confidence intervals (CIs) were used to assess the strength of the associations. We also performed genotype-based mRNA expression analysis using data from 270 individuals retrieved from public datasets. A total of 15 studies with 19796 cases and 49681 controls were included in the final meta-analysis. The pooled results revealed that the MDM4 rs4245739C allele is associated with a decreased cancer risk in the heterozygous (AC vs. AA: OR = 0.82, 95% CI = 0.73-0.93), dominant (AC/CC vs. AA: OR = 0.82, 95% CI = 0.72-0.93), and allele contrast models (C vs. A: OR = 0.84, 95% CI = 0.76-0.94). The association was more prominent in Asians and population-based studies. We also found that the rs4245739C allele was associated with decreased MDM4 mRNA expression, especially for Caucasians. Thus the MDM4 rs4245739 A > C polymorphism appears to be associated with decreased cancer risk. These findings would be strengthened by new studies with larger sample sizes and encompassing additional ethnicities.

An Approach for Predicting Essential Genes Using Multiple Homology Mapping and Machine Learning Algorithms

Investigation of essential genes is significant to comprehend the minimal gene sets of cell and discover potential drug targets. In this study, a novel approach based on multiple homology mapping and machine learning method was introduced to predict essential genes. We focused on 25 bacteria which have characterized essential genes. The predictions yielded the highest area under receiver operating characteristic (ROC) curve (AUC) of 0.9716 through tenfold cross-validation test. Proper features were utilized to construct models to make predictions in distantly related bacteria. The accuracy of predictions was evaluated via the consistency of predictions and known essential genes of target species. The highest AUC of 0.9552 and average AUC of 0.8314 were achieved when making predictions across organisms. An independent dataset from Synechococcus elongatus, which was released recently, was obtained for further assessment of the performance of our model. The AUC score of predictions is 0.7855, which is higher than other methods. This research presents that features obtained by homology mapping uniquely can achieve quite great or even better results than those integrated features. Meanwhile, the work indicates that machine learning-based method can assign more efficient weight coefficients than using empirical formula based on biological knowledge.

Pseudo progression identification of glioblastoma with dictionary learning

OBJECTIVE

Although the use of temozolomide in chemoradiotherapy is effective, the challenging clinical problem of pseudo progression has been raised in brain tumor treatment. This study aims to distinguish pseudo progression from true progression.

MATERIALS AND METHODS

Between 2000 and 2012, a total of 161 patients with glioblastoma multiforme (GBM) were treated with chemoradiotherapy at our hospital. Among the patients, 79 had their diffusion tensor imaging (DTI) data acquired at the earliest diagnosed date of pseudo progression or true progression, and 23 had both DTI data and genomic data. Clinical records of all patients were kept in good condition. Volumetric fractional anisotropy (FA) images obtained from the DTI data were decomposed into a sequence of sparse representations. Then, a feature selection algorithm was applied to extract the critical features from the feature matrix to reduce the size of the feature matrix and to improve the classification accuracy.

RESULTS

The proposed approach was validated using the 79 samples with clinical DTI data. Satisfactory results were obtained under different experimental conditions. The area under the receiver operating characteristic (ROC) curve (AUC) was 0.87 for a given dictionary with 1024 atoms. For the subgroup of 23 samples, genomics data analysis was also performed. Results implied further perspective on pseudo progression classification.

CONCLUSIONS

The proposed method can determine pseudo progression and true progression with improved accuracy. Laboring segmentation is no longer necessary because this skillfully designed method is not sensitive to tumor location.

Objective classification system for sagittal craniosynostosis based on suture segmentation

PURPOSE

Spring-assisted surgery is an effective and minimally invasive treatment for sagittal craniosynostosis (CSO). The principal barrier to the advancement of spring-assisted surgery is the patient-specific spring selection. The selection of spring force depends on the suture involved, subtypes of sagittal CSO, and age of the infant, among other factors. Clinically, physicians manually judge the subtype of sagittal CSO patients based on their CT image data, which may cause bias from different clinicians. An objective system would be helpful to stratify the sagittal CSO patients and make spring choice less subjective.

METHODS

The authors developed a novel informatics system to automatically segment and characterize sutures and classify sagittal CSO. The proposed system is composed of three phases: preprocessing, sutures segmentation, and classification. First, the three-dimensional (3D) skull was extracted from the CT images and aligned with the symmetry of the cranial vault. Second, a "hemispherical projection" algorithm was developed to transform 3D surface of the skull to a polar two-dimensional plane. Through the transformation, an "effective" projected region can be obtained to enable easy segmentation of sutures. Then, the different types of sutures, such as coronal sutures, lambdoid sutures, sagittal suture, and metopic suture, obtained from the segmented sutures were further identified by a dual-projection technique of the midline of the sutures. Finally, 108 quantified features of sutures were extracted and selected by a proposed multiclass feature scoring system. The sagittal CSO patients were classified into four subtypes: anterior, central, posterior, and complex with the support vector machine approach. Fivefold cross validation (CV) was employed to evaluate the capability of selected features in discriminating the four subtypes in 33 sagittal CSO patients. Receiver operating characteristics (ROC) curves were used to assess the robustness of the developed system.

RESULTS

The segmentation results of the proposed method were clinically acceptable for the qualitative evaluation. For the quantitative evaluation, the fivefold CV accuracy of the classification for the four subtypes was 72.7%. This classification system was reliable with the area under curve (in ROC analysis) being greater than 0.8 for four two-class problems.

CONCLUSIONS

The proposed hemispherical projection algorithm based on backtracking search can successfully segment sutures of the cranial vault. The classification system can also offer a desirable performance. As a result, the proposed segmentation and classification system is expected to bring insights into clinic research and the selection of the spring force to facilitate widespread application of this minimally invasive treatment.

Genome-wide mutational spectra analysis reveals significant cancer-specific heterogeneity

Cancer is widely recognized as a genetic disease in which somatic mutations are sequentially accumulated to drive tumor progression. Although genomic landscape studies are informative for individual cancer types, a comprehensive comparative study of tumorigenic mutations across cancer types based on integrative data sources is still a pressing need. We systematically analyzed ~10(6) non-synonymous mutations extracted from COSMIC, involving ~8000 genome-wide screened samples across 23 major human cancers at both the amino acid and gene levels. Our analysis identified cancer-specific heterogeneity that traditional nucleotide variation analysis alone usually overlooked. Particularly, the amino acid arginine (R) turns out to be the most favorable target of amino acid alteration in most cancer types studied (P < 10(-9), binomial test), reflecting its important role in cellular physiology. The tumor suppressor gene TP53 is mutated exclusively with the HYDIN, KRAS, and PTEN genes in large intestine, lung, and endometrial cancers respectively, indicating that TP53 takes part in different signaling pathways in different cancers. While some of our analyses corroborated previous observations, others indicated relevant candidates with high priority for further experimental validation. Our findings have many ramifications in understanding the etiology of cancer and the underlying molecular mechanisms in particular cancers.

Systematic Mapping of Kinase Addiction Combinations in Breast Cancer Cells by Integrating Drug Sensitivity and Selectivity Profiles

Chemical perturbation screens offer the possibility to identify actionable sets of cancer-specific vulnerabilities. However, most inhibitors of kinases or other cancer targets result in polypharmacological effects, which complicate the identification of target dependencies directly from the drug-response phenotypes. In this study, we developed a chemical systems biology approach that integrates comprehensive drug sensitivity and selectivity profiling to provide functional insights into both single and multi-target oncogenic signal addictions. When applied to 21 breast cancer cell lines, perturbed with 40 kinase inhibitors, the subtype-specific addiction patterns clustered in agreement with patient-derived subtypes, while showing considerable variability between the heterogeneous breast cancers. Experimental validation of the top predictions revealed a number of co-dependencies between kinase targets that led to unexpected synergistic combinations between their inhibitors, such as dasatinib and axitinib in the triple-negative basal-like HCC1937 cell line.

Sparse expression bases in cancer reveal tumor drivers

We define a new category of candidate tumor drivers in cancer genome evolution: ‘selected expression regulators’ (SERs)—genes driving dysregulated transcriptional programs in cancer evolution. The SERs are identified from genome-wide tumor expression data with a novel method, namely SPARROW (SPARse selected expRessiOn regulators identified With penalized regression). SPARROW uncovers a previously unknown connection between cancer expression variation and driver events, by using a novel sparse regression technique. Our results indicate that SPARROW is a powerful complementary approach to identify candidate genes containing driver events that are hard to detect from sequence data, due to a large number of passenger mutations and lack of comprehensive sequence information from a sufficiently large number of samples. SERs identified by SPARROW reveal known driver mutations in multiple human cancers, along with known cancer-associated processes and survival-associated genes, better than popular methods for inferring gene expression networks. We demonstrate that when applied to acute myeloid leukemia expression data, SPARROW identifies an apoptotic biomarker (PYCARD) for an investigational drug obatoclax. The PYCARD and obatoclax association is validated in 30 AML patient samples.

Deciphering oncogenic drivers: from single genes to integrated pathways

Technological advances in next-generation sequencing have uncovered a wide spectrum of aberrations in cancer genomes. The extreme diversity in cancer mutations necessitates computational approaches to differentiate between the 'drivers' with vital function in cancer progression and those nonfunctional 'passengers'. Although individual driver mutations are routinely identified, mutational profiles of different tumors are highly heterogeneous. There is growing consensus that pathways rather than single genes are the primary target of mutations. Here we review extant bioinformatics approaches to identifying oncogenic drivers at different mutational levels, highlighting the strategies for discovering driver pathways and networks from cancer mutation data. These approaches will help reduce the mutation complexity, thus providing a simplified picture of cancer.

Identification and analysis of driver missense mutations using rotation forest with feature selection

Identifying cancer-associated mutations (driver mutations) is critical for understanding the cellular function of cancer genome that leads to activation of oncogenes or inactivation of tumor suppressor genes. Many approaches are proposed which use supervised machine learning techniques for prediction with features obtained by some databases. However, often we do not know which features are important for driver mutations prediction. In this study, we propose a novel feature selection method (called DX) from 126 candidate features' set. In order to obtain the best performance, rotation forest algorithm was adopted to perform the experiment. On the train dataset which was collected from COSMIC and Swiss-Prot databases, we are able to obtain high prediction performance with 88.03% accuracy, 93.9% precision, and 81.35% recall when the 11 top-ranked features were used. Comparison with other various techniques in the TP53, EGFR, and Cosmic2plus datasets shows the generality of our method.

Functional annotation of putative regulatory elements at cancer susceptibility Loci

Most cancer-associated genetic variants identified from genome-wide association studies (GWAS) do not obviously change protein structure, leading to the hypothesis that the associations are attributable to regulatory polymorphisms. Translating genetic associations into mechanistic insights can be facilitated by knowledge of the causal regulatory variant (or variants) responsible for the statistical signal. Experimental validation of candidate functional variants is onerous, making bioinformatic approaches necessary to prioritize candidates for laboratory analysis. Thus, a systematic approach for recognizing functional (and, therefore, likely causal) variants in noncoding regions is an important step toward interpreting cancer risk loci. This review provides a detailed introduction to current regulatory variant annotations, followed by an overview of how to leverage these resources to prioritize candidate functional polymorphisms in regulatory regions.

Oncofuse: a computational framework for the prediction of the oncogenic potential of gene fusions

MOTIVATION

Gene fusions resulting from chromosomal aberrations are an important cause of cancer. The complexity of genomic changes in certain cancer types has hampered the identification of gene fusions by molecular cytogenetic methods, especially in carcinomas. This is changing with the advent of next-generation sequencing, which is detecting a substantial number of new fusion transcripts in individual cancer genomes. However, this poses the challenge of identifying those fusions with greater oncogenic potential amid a background of 'passenger' fusion sequences.

RESULTS

In the present work, we have used some recently identified genomic hallmarks of oncogenic fusion genes to develop a pipeline for the classification of fusion sequences, namely, Oncofuse. The pipeline predicts the oncogenic potential of novel fusion genes, calculating the probability that a fusion sequence behaves as 'driver' of the oncogenic process based on features present in known oncogenic fusions. Cross-validation and extensive validation tests on independent datasets suggest a robust behavior with good precision and recall rates. We believe that Oncofuse could become a useful tool to guide experimental validation studies of novel fusion sequences found during next-generation sequencing analysis of cancer transcriptomes.

AVAILABILITY AND IMPLEMENTATION

Oncofuse is a naive Bayes Network Classifier trained and tested using Weka machine learning package. The pipeline is executed by running a Java/Groovy script, available for download at www.unav.es/genetica/oncofuse.html.