Gene expression profile based classification models of psoriasis

Prediction of crucial epigenetically‑associated, differentially expressed genes by integrated bioinformatics analysis and the identification of S100A9 as a novel biomarker in psoriasis

Psoriasis is one of the most common immune-mediated inflammatory diseases of the skin. The identification of the pivotal molecular mechanisms responsible for the disease pathogenesis may lead to the development of novel therapeutic options. The present study aimed to identify pivotal differentially expressed genes (DEGs) and methylated DEGs in psoriasis. The raw data from gene microarrays were obtained from the Gene Expression Omnibus database. The data were processed using packages in Bioconductor. In total, 352 upregulated and 137 downregulated DEGs were identified. The upregulated DEGs were primarily enriched in the 'innate immune defense' response and the 'cell cycle'. The down-regulated DEGs were primarily enriched in 'cell adhesion' and 'tight junction pathways'. A total of 95 methylated DEGs were identified, which were significantly enriched in the 'interleukin (IL)-17 signaling pathway' and the 'response to interferon'. Based on a comprehensive evaluation of all algorithms in cytoHubba, the key epigenetic-associated hub genes (S100A9, SELL, FCGR3B, MMP9, S100A7, IL7R, IRF7, CCR7, IFI44, CXCL1 and LCN2) were screened out. In order to further validate these genes, the present study constructed a model of imiquimod (IMQ)-induced psoriasiform dermatitis using mice. The levels of these hub genes were increased in the IMQ group. The knockdown of methylation-regulating enzyme ten-eleven translocation (TET) 2 expression in mice attenuated the expression levels of S100A9, SELL, IL7R, MMP9, CXCL1 and LCN2. Furthermore, the hydroxymethylated level of S100A9 was highly expressed in the IMQ group and was significantly decreased by TET2 deficiency in mice. On the whole, using an integrative system bioinformatics approach, the present study identified a series of characteristic enrichment pathways and key genes that may serve as potential biomarkers in psoriasis.

A nomogram based on a gene signature for predicting the prognosis of patients with head and neck squamous cell carcinoma

BACKGROUND

Head and neck squamous cell carcinoma (HNSCC) is one of the most common malignant tumors. The purpose of this study was to establish and validate a gene-expression-based prognostic signature in non-metastatic patients with HNSCC.

MATERIALS AND METHODS

All the patients were retrieved from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. We randomly divided the GSE65858 samples into 70% (training cohort, n = 190) and 30% (internal validation cohort, n = 72). A total of 36 samples collected from the TCGA HNSCC databases were selected as an independent external validation cohort. The oligo package in R was used to normalize the raw data before analysis. Data characteristics were extracted, and a gene signature was built via the least absolute shrinkage and selection operator regression model. The predictive model was developed by multivariable Cox regression analysis. T stage, N stage, human papilloma virus status, and the gene signature were incorporated in this predictive model, which was shown as a nomogram. Calibration and discrimination were performed to assess the performance of the nomogram. The clinical utility of this nomogram was assessed by the decision curve analysis.

RESULTS

Overall, 2001 significant messenger RNAs in HNSCC samples were identified compared with normal samples. The gene signature contained seven genes and significantly correlated with overall survival. The gene signature was also significant in subgroup analysis of the primary cohort. The calibration was plotted in the external cohort (C-index 0.90, 95% CI 0.85, 0.95) compared with the training (C-index 0.76, 95% CI 0.73, 0.79) and internal (C-index 0.71, 95% CI 0.66, 0.77) cohorts. In clinic, a decision curve analysis demonstrated that the model including the prognostic gene signature score status was better than that without it.

CONCLUSION

This study developed and validated a predictive model, which can promote the individualized prediction of overall survival in non-metastatic patients with HNSCC.

Silencing KRT16 inhibits keratinocyte proliferation and VEGF secretion in psoriasis via inhibition of ERK signaling pathway

Psoriasis is a multisystem disease affecting about 2% of the population, while keratin16 (KRT16) has been reported to participate in psoriasis. However, the specific mechanism of KRT16 in psoriasis was inadequately investigated. The objective of the study was to elucidate the mechanism by which siRNA-mediated silencing of KRT16 affects keratinocyte proliferation and vascular endothelial growth factor (VEGF) secretion in psoriasis through the extracellular signal-related kinase (ERK) signaling pathway. Psoriasis-related core gene KRT16 was screened out. Then, the expression of KRT16, VEGF, and ERK signaling pathway-related genes was detected in psoriatic patients. To further investigate the mechanism of KRT16, keratinocytes in psoriatic patients were treated with KRT16 siRNA or/and ERK inhibitor (PD98059) to detect the changes in related gene expression and cell survival. KRT16 was involved in psoriasis development. The expression levels of KRT16, p-ERK1/2, and VEGF in lesion tissues are significantly elevated. Keratinocytes treated with KRT16-siRNA and KRT16-siRNA + PD98059 exhibited reduced KRT16, p-ERK1/2, and VEGF expression. The cell survival rate in cells treated with KRT16-siRNA, PD98059, and KRT16-siRNA + PD98059 reduced significantly. These findings indicate that silencing KRT16 inhibits keratinocyte proliferation and VEGF secretion in psoriasis via inhibition of ERK signaling pathway, which provides a basic theory in the treatment of psoriasis.

Evaluation of variable selection methods for random forests and omics data sets

Machine learning methods and in particular random forests are promising approaches for prediction based on high dimensional omics data sets. They provide variable importance measures to rank predictors according to their predictive power. If building a prediction model is the main goal of a study, often a minimal set of variables with good prediction performance is selected. However, if the objective is the identification of involved variables to find active networks and pathways, approaches that aim to select all relevant variables should be preferred. We evaluated several variable selection procedures based on simulated data as well as publicly available experimental methylation and gene expression data. Our comparison included the Boruta algorithm, the Vita method, recurrent relative variable importance, a permutation approach and its parametric variant (Altmann) as well as recursive feature elimination (RFE).  In our simulation studies, Boruta was the most powerful approach, followed closely by the Vita method. Both approaches demonstrated similar stability in variable selection, while Vita was the most robust approach under a pure null model without any predictor variables related to the outcome. In the analysis of the different experimental data sets, Vita demonstrated slightly better stability in variable selection and was less computationally intensive than Boruta. In conclusion, we recommend the Boruta and Vita approaches for the analysis of high-dimensional data sets. Vita is considerably faster than Boruta and thus more suitable for large data sets, but only Boruta can also be applied in low-dimensional settings.

A Controlled Study of the Feasibility and Efficacy of a Cloud-Based Interactive Management Program Between Patients with Psoriasis and Physicians

BACKGROUND Dermatology patients continue to receive improved treatment, but for patients with psoriasis, there have been few studies on ways to improve patient management by improving communication with patients and their dermatologists. This study aimed to investigate the feasibility and efficacy of cloud-based interactive patient and physician management of psoriasis. MATERIAL AND METHODS The cloud-based platform was created by professional software engineers to educate and manage patients with psoriasis in a single hospital, where patients and research staff had a network platform for sharing data. A total of 79 patients with psoriasis were included in this study and were randomly divided into the control group (n=39) and the intervention group (n=40). Patients in the control group were given a psoriasis nursing manual and underwent regular follow-up. Patients in the intervention group were managed using the cloud platform, with the same management as the control group. The Psoriasis Area Severity Index (PASI), the Self-Rating Anxiety Scale (SAS), the Dermatology Life Quality Index (DLQI), and the Symptom Checklist-90-Revised (SCL-90-R) were used. RESULTS Cloud-based interactive patient and physician management resulted in clinical improvement, and reduced the degree of anxiety in patients with psoriasis and improved their physical and mental health. Patients in the intervention group had an improved understanding of psoriasis treatment, resulting in an improved relationship with the medical staff and improved treatment compliance. CONCLUSIONS Cloud-based interactive patient and physician management improved the mental health and quality of life for patients with psoriasis and allowed patients to manage their disease more effectively.

DNA methylation-based subclassification of psoriasis in the Chinese Han population

Psoriasis (Ps) is an inflammatory skin disease caused by genetic and environmental factors. Previous studies on DNA methylation (DNAm) found genetic markers that are closely associated with Ps, and evidence has shown that DNAm mediates genetic risk in Ps. In this study, Consensus Clustering was used to analyze DNAm data, and 114 Ps patients were divided into three subclassifications. Investigation of the clinical characteristics and copy number variations (CNVs) of DEFB4, IL22, and LCE3C in the three subclassifications revealed no significant differences in gender ratio and in Ps area and severity index (PASI) score. The proportion of late-onset ( ≥ 40 years) Ps patients was significantly higher in type I than in types II and III (P = 0.035). Type III contained the smallest proportion of smokers and the largest proportion of non-smoking Ps patients (P = 0.086). The CNVs of DEFB4 and LCE3C showed no significant differences but the CNV of IL22 significantly differed among the three subclassifications (P = 0.044). This study is the first to profile Ps subclassifications based on DNAm data in the Chinese Han population. These results are useful in the treatment and management of Ps from the molecular and genetic perspectives.

Clinical and molecular evaluation of therapy with the use of cyclosporine A in patients with psoriasis vulgaris

BACKGROUND

Psoriasis course involves increased secretion of pro-inflammatory cytokines, among others, a beta transforming growth factor (TGFβs) and its receptors. Cyclosporine A (CsA), an immunosuppressive medicine with the molecular mechanism of operation connected with the properties of cell cycle suppression, is often used in the treatment of severe forms of psoriasis. The efficacy of therapy is assessed based on the disease clinical progression indexes - Psoriasis Area and Severity Index (PASI), body surface area (BSA), and Dermatology Life Quality Index (DLQI). The aim of the study was the evaluation of the efficacy of the CsA treatment of patients with psoriasis vulgaris, based on the clinical parameters and an assessment of the expression profiles of TGFβs and TGFβRs, depending on the concurrent diabetes and metabolic syndrome.

METHODS

The group under study composed of 32 patients (15 with the metabolic syndrome, seven with diabetes) treated with CsA for 84 days. The molecular analysis included extraction of RNA, assessment of TGβF1-3, TGFβRI-III gene expression with the use of the RTqPCR method. The clinical assessment of the effects of this pharmacotherapy involved evaluation of the parameters: PASI, BSA, DLQI before therapy commencement, on the 42nd and 84th days of therapy.

RESULTS

A statistically significant change in the transcription activity of TGFβ1 in patients with and without diabetes (P = 0.018) and patients with and without metabolic syndrome (P = 0.023) was shown that on the 84th day of therapy.

CONCLUSIONS

TGFb1 may be claimed as the supplementary molecular marker to evaluate the efficacy of CsA therapy. It seems that systemic diseases have an effect on the efficacy of the applied pharmacotherapy and the course of psoriasis.

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Biomarker detection is one of the more important biomedical questions for high-throughput 'omics' researchers, and almost all existing biomarker detection algorithms generate one biomarker subset with the optimized performance measurement for a given dataset. However, a recent study demonstrated the existence of multiple biomarker subsets with similarly effective or even identical classification performances. This protocol presents a simple and straightforward methodology for detecting biomarker subsets with binary classification performances, better than a user-defined cutoff. The protocol consists of data preparation and loading, baseline information summarization, parameter tuning, biomarker screening, result visualization and interpretation, biomarker gene annotations, and result and visualization exportation at publication quality. The proposed biomarker screening strategy is intuitive and demonstrates a general rule for developing biomarker detection algorithms. A user-friendly graphical user interface (GUI) was developed using the programming language Python, allowing biomedical researchers to have direct access to their results. The source code and manual of kSolutionVis can be downloaded from http://www.healthinformaticslab.org/supp/resources.php.

A powerful method to integrate genotype and gene expression data for dissecting the genetic architecture of a disease

To decipher the genetic architecture of human disease, various types of omics data are generated. Two common omics data are genotypes and gene expression. Often genotype data for a large number of individuals and gene expression data for a few individuals are generated due to biological and technical reasons, leading to unequal sample sizes for different omics data. Unavailability of standard statistical procedure for integrating such datasets motivates us to propose a two-step multi-locus association method using latent variables. Our method is powerful than single/separate omics data analysis and it unravels comprehensively deep-seated signals through a single statistical model. Extensive simulation confirms that it is robust to various genetic models as its power increases with sample size and number of associated loci. It provides p-values very fast. Application to real dataset on psoriasis identifies 17 novel SNPs, functionally related to psoriasis-associated genes, at much smaller sample size than standard GWAS.

Variances in the mRNA expression profile of TGF-β1-3 isoforms and its TGF-βRI-III receptors during cyclosporin a treatment of psoriatic patients

Introduction

Psoriasis is a chronic, immunologic, multi-factor inflammatory skin disease, strongly associated with a higher level of a number of cytokines, such as isoforms of transforming growth factor β (TGF-β1–3) and its receptors (TGF-βRI–III). One of the most popular and important drugs used to treat this disease is cyclosporin A (CsA).

Aim

The aim of this study was to investigate the expression of genes encoding the transforming growth factor (TGF)-β isoforms and receptors of the cytokine TGF-βRs in psoriatic patients during an 84-day long observation of the effects of cyclosporin A therapy. It made an attempt to determine the usefulness of testing mRNA expression of TGF-β1–3 and its receptors TGF-βRI–III as the supplementary molecular markers of lost sensitivity to the medicine.

Material and methods

The study group consisted of 32 patients with psoriasis (20 men and 12 women) treated with cyclosporin A. The changes in expression patterns of TGF-β1-3 and TGF-βRI-III were performed by real-time quantitative reverse transcription PCR (RTqPCR).

Results

The expression of TGF-β1-3 and TGF-βRI-III were detected in the whole period of therapy with CsA. Changes in transcriptional activities of TGF-β1–3 and TGF-βRI–III during pharmacotherapy were observed. Differences in the expression of these genes were found before and after 42 and 84 days of using CsA.

Conclusions

The changes in expression profiles of TGF-β1-3 and TGF-βRI-III during CsA therapy can be a useful molecular marker of lost sensitivity to the medicine.

C10orf99 contributes to the development of psoriasis by promoting the proliferation of keratinocytes

Psoriasis is a chronic, relapsing inflammatory skin disease. The pathogenesis of psoriasis is complex and has not been fully understood. C10orf99 was a recently identified human antimicrobial peptide whose mRNA expression is elevated in psoriatic human skin samples. In this study, we investigated the functional roles of C10orf99 in epidermal proliferation under inflammatory condition. We showed that C10orf99 protein was significantly up-regulated in psoriatic skin samples from patients and the ortholog gene expression levels were up-regulated in imiquimod (IMQ)-induced psoriasis-like skin lesions in mice. Using M5-stimulated HaCaT cell line model of inflammation and a combinational approach of knockdown and overexpression of C10orf99, we demonstrated that C10orf99 could promote keratinocyte proliferation by facilitating the G1/S transition, and the pro-proliferation effect of C10orf99 was associated with the activation of the ERK1/2 and NF-κB but not the AKT pathways. Local depletion of C10orf99 by lentiviral vectors expressing C10orf99 shRNA effectively ameliorated IMQ-induced dermatitis. Taken together, these results indicate that C10orf99 plays a contributive role in psoriasis pathogenesis and may serve as a new target for psoriasis treatment.

Gender specificity improves the early-stage detection of clear cell renal cell carcinoma based on methylomic biomarkers

AIM

The two genders are different ranging from the molecular to the phenotypic levels. But most studies did not use this important information. We hypothesize that the integration of gender information may improve the overall prediction accuracy.

MATERIALS & METHODS

A comprehensive comparative study was carried out to test the hypothesis. The classification of the stages I + II versus III + IV of the clear cell renal cell carcinoma samples was formulated as an example.

RESULTS & CONCLUSION

In most cases, female-specific model significantly outperformed both-gender model, as similarly for the male-specific model. Our data suggested that gender information is essential for building biomedical classification models and even a simple strategy of building two gender-specific models may outperform the gender-mixed model.

An OMIC biomarker detection algorithm TriVote and its application in methylomic biomarker detection

Aim: Transcriptomic and methylomic patterns represent two major OMIC data sources impacted by both inheritable genetic information and environmental factors, and have been widely used as disease diagnosis and prognosis biomarkers. Materials & methods: Modern transcriptomic and methylomic profiling technologies detect the status of tens of thousands or even millions of probing residues in the human genome, and introduce a major computational challenge for the existing feature selection algorithms. This study proposes a three-step feature selection algorithm, TriVote, to detect a subset of transcriptomic or methylomic residues with highly accurate binary classification performance. Results & conclusion: TriVote outperforms both filter and wrapper feature selection algorithms with both higher classification accuracy and smaller feature number on 17 transcriptomes and two methylomes. Biological functions of the methylome biomarkers detected by TriVote were discussed for their disease associations. An easy-to-use Python package is also released to facilitate the further applications.

Multiregional radiomics features from multiparametric MRI for prediction of MGMT methylation status in glioblastoma multiforme: A multicentre study

OBJECTIVES

To build a reliable radiomics model from multiregional and multiparametric magnetic resonance imaging (MRI) for pretreatment prediction of O⁶-methylguanine-DNA methyltransferase (MGMT) promotor methylation status in glioblastoma multiforme (GBM).

METHODS

In this retrospective multicentre study, 1,705 multiregional radiomics features were automatically extracted from multiparametric MRI. A radiomics model with a minimal set of all-relevant features and a radiomics model with univariately-predictive and non-redundant features were built for MGMT methylation prediction from a primary cohort (133 patients) and tested on an independent validation cohort (60 patients). Predictive models combing clinical factors were built and evaluated. Both radiomics models were assessed on subgroups stratified by clinical factors.

RESULTS

The radiomics model with six all-relevant features allowed pretreatment prediction of MGMT methylation (AUC=0.88, accuracy=80 %), which significantly outperformed the model with eight univariately-predictive and non-redundant features (AUC=0.76, accuracy=70 %). Combing clinical factors with radiomics features did not benefit the prediction performance. The all-relevant model achieved significantly better performance in stratified analysis.

CONCLUSIONS

Radiomics model built from multiregional and multiparameter MRI may serve as a potential imaging biomarker for pretreatment prediction of MGMT methylation in GBM. The all-relevant features have the potential of offering better predictive power than the univariately-predictive and non-redundant features.

KEY POINTS

• Multiregional and multiparametric MRI features reliably predicted MGMT methylation in multicentre cohorts. • All-relevant imaging features predicted MGMT methylation better than univariately-predictive and non-redundant features. • Combing clinical factors with radiomics features did not benefit the prediction performance.

Molecular action of isoflavone genistein in the human epithelial cell line HaCaT

Due to its strong proliferation-reducing effects on keratinocytes, and also anti-inflammatory properties, the isoflavone genistein has already been proposed as a possible antipsoriatic compound. As there is still no detailed information on this topic, we examined the effects of genistein by using an in vitro model of both, normal and "psoriasis-like" keratinocytes at this stage of our work exhaustively testing the selected flavonoid in a mono-treated experimental design. Gene expression studies revealed transcriptional changes that confirms known disease-associated pathways and highlights many psoriasis-related genes. Our results suggested that aberrant expression of genes contributing to the progress of psoriasis could be improved by the action of genistein. Genistein prevented "cytokine mix" as well as TNF-α-induced NF-κB nuclear translocation, with no effect on the PI3K signaling cascade, indicating the luck of turning this pathway into NF-κB activation. It could have attenuated TNF-α and LPS-induced inflammatory responses by suppressing ROS activation. Regardless of the type of keratinocyte stimulation used, reduction of cytokine IL-8, IL-20 and CCL2 production (both at RNA and protein level) following genistein treatment was visible. Because investigations of other groups supported our commentary on potential administration of genistein as a potential weapon in the armamentarium against psoriasis, it is believed that this paper should serve to encourage researchers to conduct further studies on this subject.

Models in the Research Process of Psoriasis

Psoriasis is an ancient, universal chronic skin disease with a significant geographical variability, with the lowest incidence rate at the equator, increasing towards the poles. Insights into the mechanisms responsible for psoriasis have generated an increasing number of druggable targets and molecular drugs. The development of relevant in vitro and in vivo models of psoriasis is now a priority and an important step towards its cure. In this review, we summarize the current cellular and animal systems suited to the study of psoriasis. We discuss the strengths and limitations of the various models and the lessons learned. We conclude that, so far, there is no one model that can meet all of the research needs. Therefore, the choice model system will depend on the questions being addressed.

RIFS: a randomly restarted incremental feature selection algorithm

The advent of big data era has imposed both running time and learning efficiency challenges for the machine learning researchers. Biomedical OMIC research is one of these big data areas and has changed the biomedical research drastically. But the high cost of data production and difficulty in participant recruitment introduce the paradigm of “large p small n” into the biomedical research. Feature selection is usually employed to reduce the high number of biomedical features, so that a stable data-independent classification or regression model may be achieved. This study randomly changes the first element of the widely-used incremental feature selection (IFS) strategy and selects the best feature subset that may be ranked low by the statistical association evaluation algorithms, e.g. t-test. The hypothesis is that two low-ranked features may be orchestrated to achieve a good classification performance. The proposed Randomly re-started Incremental Feature Selection (RIFS) algorithm demonstrates both higher classification accuracy and smaller feature number than the existing algorithms. RIFS also outperforms the existing methylomic diagnosis model for the prostate malignancy with a larger accuracy and a lower number of transcriptomic features.

A Mucosal and Cutaneous Chemokine Ligand for the Lymphocyte Chemoattractant Receptor GPR15

Chemoattractants control lymphocyte recruitment from the blood, contributing to the systemic organization of the immune system. The G protein-linked receptor GPR15 mediates lymphocyte homing to the large intestines and skin. Here we show that the 9 kDa CC-motif containing cationic polypeptide AP57/colon-derived sushi containing domain-2 binding factor (CSBF), encoded by C10orf99 in the human and 2610528A11Rik in the mouse, functions as a chemokine ligand for GPR15 (GPR15L). GPR15L binds GPR15 and attracts GPR15-expressing T cells including lymphocytes in colon-draining lymph nodes and Vγ3⁺ thymic precursors of dermal epithelial T cells. Patterns of GPR15L expression by epithelial cells in adult mice and humans suggest a homeostatic role for the chemokine in lymphocyte localization to the large intestines, as well as a role in homing to the epidermis during wound healing or inflammation. GPR15L is also significantly expressed in squamous mucosa of the oral cavity and esophagus with still poorly defined regulation. Identification of the chemotactic activity of GPR15L adds to its reported antibacterial and tumor cell growth regulatory functions and suggests the potential of targeting GPR15L–GPR15 interactions for modulation of mucosal and cutaneous inflammation.

hMuLab: A Biomedical Hybrid MUlti-LABel Classifier Based on Multiple Linear Regression

Many biomedical classification problems are multi-label by nature, e.g., a gene involved in a variety of functions and a patient with multiple diseases. The majority of existing classification algorithms assumes each sample with only one class label, and the multi-label classification problem remains to be a challenge for biomedical researchers. This study proposes a novel multi-label learning algorithm, hMuLab, by integrating both feature-based and neighbor-based similarity scores. The multiple linear regression modeling techniques make hMuLab capable of producing multiple label assignments for a query sample. The comparison results over six commonly-used multi-label performance measurements suggest that hMuLab performs accurately and stably for the biomedical datasets, and may serve as a complement to the existing literature.

Gene selection for tumor classification using a novel bio-inspired multi-objective approach

Identifying the informative genes has always been a major step in microarray data analysis. The complexity of various cancer datasets makes this issue still challenging. In this paper, a novel Bio-inspired Multi-objective algorithm is proposed for gene selection in microarray data classification specifically in the binary domain of feature selection. The presented method extends the traditional Bat Algorithm with refined formulations, effective multi-objective operators, and novel local search strategies employing social learning concepts in designing random walks. A hybrid model using the Fisher criterion is then applied to three widely-used microarray cancer datasets to explore significant biomarkers which reveal the effectiveness of the proposed method for genomic analysis. Experimental results unveil new combinations of informative biomarkers have association with other studies.

Monitoring seasonal influenza epidemics by using internet search data with an ensemble penalized regression model

Seasonal influenza epidemics cause serious public health problems in China. Search queries-based surveillance was recently proposed to complement traditional monitoring approaches of influenza epidemics. However, developing robust techniques of search query selection and enhancing predictability for influenza epidemics remains a challenge. This study aimed to develop a novel ensemble framework to improve penalized regression models for detecting influenza epidemics by using Baidu search engine query data from China. The ensemble framework applied a combination of bootstrap aggregating (bagging) and rank aggregation method to optimize penalized regression models. Different algorithms including lasso, ridge, elastic net and the algorithms in the proposed ensemble framework were compared by using Baidu search engine queries. Most of the selected search terms captured the peaks and troughs of the time series curves of influenza cases. The predictability of the conventional penalized regression models were improved by the proposed ensemble framework. The elastic net regression model outperformed the compared models, with the minimum prediction errors. We established a Baidu search engine queries-based surveillance model for monitoring influenza epidemics, and the proposed model provides a useful tool to support the public health response to influenza and other infectious diseases.

Gene selection for microarray cancer classification using a new evolutionary method employing artificial intelligence concepts

Gene selection is a demanding task for microarray data analysis. The diverse complexity of different cancers makes this issue still challenging. In this study, a novel evolutionary method based on genetic algorithms and artificial intelligence is proposed to identify predictive genes for cancer classification. A filter method was first applied to reduce the dimensionality of feature space followed by employing an integer-coded genetic algorithm with dynamic-length genotype, intelligent parameter settings, and modified operators. The algorithmic behaviors including convergence trends, mutation and crossover rate changes, and running time were studied, conceptually discussed, and shown to be coherent with literature findings. Two well-known filter methods, Laplacian and Fisher score, were examined considering similarities, the quality of selected genes, and their influences on the evolutionary approach. Several statistical tests concerning choice of classifier, choice of dataset, and choice of filter method were performed, and they revealed some significant differences between the performance of different classifiers and filter methods over datasets. The proposed method was benchmarked upon five popular high-dimensional cancer datasets; for each, top explored genes were reported. Comparing the experimental results with several state-of-the-art methods revealed that the proposed method outperforms previous methods in DLBCL dataset.

CpG Methylation Signature Predicts Recurrence in Early-Stage Hepatocellular Carcinoma: Results From a Multicenter Study

Purpose Early-stage hepatocellular carcinoma (E-HCC) is being diagnosed increasingly, and in one half of diagnosed patients, recurrence will develop. Thus, it is urgent to identify recurrence-related markers. We investigated the effectiveness of CpG methylation in predicting recurrence for patients with E-HCCs. Patients and Methods In total, 576 patients with E-HCC from four independent centers were sorted by three phases. In the discovery phase, 66 tumor samples were analyzed using the Illumina Methylation 450k Beadchip. Two algorithms, Least Absolute Shrinkage and Selector Operation and Support Vector Machine-Recursive Feature Elimination, were used to select significant CpGs. In the training phase, penalized Cox regression was used to further narrow CpGs into 140 samples. In the validation phase, candidate CpGs were validated using an internal cohort (n = 141) and two external cohorts (n = 191 and n =104). Results After combining the 46 CpGs selected by the Least Absolute Shrinkage and Selector Operation and the Support Vector Machine-Recursive Feature Elimination algorithms, three CpGs corresponding to SCAN domain containing 3, Src homology 3-domain growth factor receptor-bound 2-like interacting protein 1, and peptidase inhibitor 3 were highlighted as candidate predictors in the training phase. On the basis of the three CpGs, a methylation signature for E-HCC (MSEH) was developed to classify patients into high- and low-risk recurrence groups in the training cohort ( P < .001). The performance of MSEH was validated in the internal cohort ( P < .001) and in the two external cohorts ( P < .001; P = .002). Furthermore, a nomogram comprising MSEH, tumor differentiation, cirrhosis, hepatitis B virus surface antigen, and antivirus therapy was generated to predict the 5-year recurrence-free survival in the training cohort, and it performed well in the three validation cohorts (concordance index: 0.725, 0.697, and 0.693, respectively). Conclusion MSEH, a three-CpG-based signature, is useful in predicting recurrence for patients with E-HCC.

CRISPRdigger: detecting CRISPRs with better direct repeat annotations

Clustered regularly interspaced short palindromic repeats (CRISPRs) are important genetic elements in many bacterial and archaeal genomes, and play a key role in prokaryote immune systems’ fight against invasive foreign elements. The CRISPR system has also been engineered to facilitate target gene editing in eukaryotic genomes. Using the common features of mis-annotated CRISPRs in prokaryotic genomes, this study proposed an accurate de novo CRISPR annotation program CRISPRdigger, which can take a partially assembled genome as its input. A comprehensive comparison with the three existing programs demonstrated that CRISPRdigger can recover more Direct Repeats (DRs) for CRISPRs and achieve a higher accuracy for a query genome. The program was implemented by Perl and all the parameters had default values, so that a user could annotate CRISPRs in a query genome by supplying only a genome sequence in the FASTA format. All the supplementary data are available at http://www.healthinformaticslab.org/supp/.

CD3D and PRKCQ work together to discriminate between B-cell and T-cell acute lymphoblastic leukemia

Different therapeutic methods have been developed for the B-cell and T-cell subtypes of acute lymphoblastic leukemia (ALL). The identification of molecular biomarkers that can accurately discriminate between B-cell and T-cell ALLs will facilitate the quick determination of therapeutic plans, as well as reveal the intrinsic mechanisms underlining the two different ALL subtypes. This study computationally screened the high-throughput transcriptome dataset for multiple candidate biomarkers and verified their discrimination abilities in an independent sample set using quantitative real-time polymerase chain reaction (PCR) technology. Both technologies suggest that the two genes CD3D and PKRCQ together provided a good model for classification of B-cell and T-cell ALLs, whereas the individual genes did not show consistent discrimination between the two ALL subtypes. Supplementary material is available at http://healthinformaticslab.org/supp/.

Filtered selection coupled with support vector machines generate a functionally relevant prediction model for colorectal cancer

Purpose

There has been considerable interest in using whole-genome expression profiles for the classification of colorectal cancer (CRC). The selection of important features is a crucial step before training a classifier.

Methods

In this study, we built a model that uses support vector machine (SVM) to classify cancer and normal samples using Affymetrix exon microarray data obtained from 90 samples of 48 patients diagnosed with CRC. From the 22,011 genes, we selected the 20, 30, 50, 100, 200, 300, and 500 genes most relevant to CRC using the minimum-redundancy–maximum-relevance (mRMR) technique. With these gene sets, an SVM model was designed using four different kernel types (linear, polynomial, radial basis function [RBF], and sigmoid).

Results

The best model, which used 30 genes and RBF kernel, outperformed other combinations; it had an accuracy of 84% for both ten fold and leave-one-out cross validations in discriminating the cancer samples from the normal samples. With this 30 genes set from mRMR, six classifiers were trained using random forest (RF), Bayes net (BN), multilayer perceptron (MLP), naïve Bayes (NB), reduced error pruning tree (REPT), and SVM. Two hybrids, mRMR + SVM and mRMR + BN, were the best models when tested on other datasets, and they achieved a prediction accuracy of 95.27% and 91.99%, respectively, compared to other mRMR hybrid models (mRMR + RF, mRMR + NB, mRMR + REPT, and mRMR + MLP). Ingenuity pathway analysis was used to analyze the functions of the 30 genes selected for this model and their potential association with CRC: CDH3, CEACAM7, CLDN1, IL8, IL6R, MMP1, MMP7, and TGFB1 were predicted to be CRC biomarkers.

Conclusion

This model could be used to further develop a diagnostic tool for predicting CRC based on gene expression data from patient samples.

Computer-aided diagnosis of psoriasis skin images with HOS, texture and color features: A first comparative study of its kind

Psoriasis is an autoimmune skin disease with red and scaly plaques on skin and affecting about 125 million people worldwide. Currently, dermatologist use visual and haptic methods for diagnosis the disease severity. This does not help them in stratification and risk assessment of the lesion stage and grade. Further, current methods add complexity during monitoring and follow-up phase. The current diagnostic tools lead to subjectivity in decision making and are unreliable and laborious. This paper presents a first comparative performance study of its kind using principal component analysis (PCA) based CADx system for psoriasis risk stratification and image classification utilizing: (i) 11 higher order spectra (HOS) features, (ii) 60 texture features, and (iii) 86 color feature sets and their seven combinations. Aggregate 540 image samples (270 healthy and 270 diseased) from 30 psoriasis patients of Indian ethnic origin are used in our database. Machine learning using PCA is used for dominant feature selection which is then fed to support vector machine classifier (SVM) to obtain optimized performance. Three different protocols are implemented using three kinds of feature sets. Reliability index of the CADx is computed. Among all feature combinations, the CADx system shows optimal performance of 100% accuracy, 100% sensitivity and specificity, when all three sets of feature are combined. Further, our experimental result with increasing data size shows that all feature combinations yield high reliability index throughout the PCA-cutoffs except color feature set and combination of color and texture feature sets. HOS features are powerful in psoriasis disease classification and stratification. Even though, independently, all three set of features HOS, texture, and color perform competitively, but when combined, the machine learning system performs the best. The system is fully automated, reliable and accurate.

McTwo: a two-step feature selection algorithm based on maximal information coefficient

BACKGROUND

High-throughput bio-OMIC technologies are producing high-dimension data from bio-samples at an ever increasing rate, whereas the training sample number in a traditional experiment remains small due to various difficulties. This "large p, small n" paradigm in the area of biomedical "big data" may be at least partly solved by feature selection algorithms, which select only features significantly associated with phenotypes. Feature selection is an NP-hard problem. Due to the exponentially increased time requirement for finding the globally optimal solution, all the existing feature selection algorithms employ heuristic rules to find locally optimal solutions, and their solutions achieve different performances on different datasets.

RESULTS

This work describes a feature selection algorithm based on a recently published correlation measurement, Maximal Information Coefficient (MIC). The proposed algorithm, McTwo, aims to select features associated with phenotypes, independently of each other, and achieving high classification performance of the nearest neighbor algorithm. Based on the comparative study of 17 datasets, McTwo performs about as well as or better than existing algorithms, with significantly reduced numbers of selected features. The features selected by McTwo also appear to have particular biomedical relevance to the phenotypes from the literature.

CONCLUSION

McTwo selects a feature subset with very good classification performance, as well as a small feature number. So McTwo may represent a complementary feature selection algorithm for the high-dimensional biomedical datasets.

Computational Systems Biology of Psoriasis: Are We Ready for the Age of Omics and Systems Biomarkers?

Computational biology and 'omics' systems sciences are greatly impacting research on common diseases such as cancer. By contrast, dermatology covering an array of skin diseases with high prevalence in society, has received relatively less attention from 'omics' and computational biosciences. We are focusing on psoriasis, a common and debilitating autoimmune disease involving skin and joints. Using computational systems biology and reconstruction, topological, modular, and a novel correlational analyses (based on fold changes) of biological and transcriptional regulatory networks, we analyzed and integrated data from a total of twelve studies from the Gene Expression Omnibus (sample size = 534). Samples represented a comprehensive continuum from lesional and nonlesional skin, as well as bone marrow and dermal mesenchymal stem cells. We identified and propose here a JAK/STAT signaling pathway significant for psoriasis. Importantly, cytokines, interferon-stimulated genes, antimicrobial peptides, among other proteins, were involved in intrinsic parts of the proposed pathway. Several biomarker and therapeutic candidates such as SUB1 are discussed for future experimental studies. The integrative systems biology approach presented here illustrates a comprehensive perspective on the molecular basis of psoriasis. This also attests to the promise of systems biology research in skin diseases, with psoriasis as a systemic component. The present study reports, to the best of our knowledge, the largest set of microarray datasets on psoriasis, to offer new insights into the disease mechanisms with a proposal of a disease pathway. We call for greater computational systems biology research and analyses in dermatology and skin diseases in general.

In situ gel-forming AP-57 peptide delivery system for cutaneous wound healing

In situ gel-forming system as local drug delivery system in dermal traumas has generated a great interest. Accumulating evidence shows that antimicrobial peptides play pivotal roles in the process of wound healing. Here in this study, to explore the potential application of antimicrobial peptide in wound healing, biodegradable poly(L-lactic acid)-Pluronic L35-poly(L-lactic acid) (PLLA-L35-PLLA) was developed at first. Then based on this polymer, an injectable in situ gel-forming system composed of human antimicrobial peptides 57 (AP-57) loaded nanoparticles and thermosensitive hydrogel was prepared and applied for cutaneous wound healing. AP-57 peptides were enclosed with biocompatible nanoparticles (AP-57-NPs) with high drug loading and encapsulation efficiency. AP-57-NPs were further encapsulated in a thermosensitive hydrogel (AP-57-NPs-H) to facilitate its application in cutaneous wound repair. As a result, AP-57-NPs-H released AP-57 in an extended period and exhibited quite low cytotoxicity and high anti-oxidant activity in vitro. Moreover, AP-57-NPs-H was free-flowing liquid at room temperature, and can form non-flowing gel without any crosslink agent upon applied on the wounds. In vivo wound healing assay using full-thickness dermal defect model of SD rats indicated that AP-57-NPs-H could significantly promote wound healing. At day 14 after operation, AP-57-NPs-H treated group showed nearly complete wound closure of 96.78 ± 3.12%, whereas NS, NPs-H and AP-57-NPs group recovered by about 68.78 ± 4.93%, 81.96 ± 3.26% and 87.80 ± 4.62%, respectively. Histopathological examination suggested that AP-57-NPs-H could promote cutaneous wound healing through enhancing granulation tissue formation, increasing collagen deposition and promoting angiogenesis in the wound tissue. Therefore, AP-57-NPs-H might have potential application in wound healing.

Improved Variable Selection Algorithm Using a LASSO-Type Penalty, with an Application to Assessing Hepatitis B Infection Relevant Factors in Community Residents

OBJECTIVES

In epidemiological studies, it is important to identify independent associations between collective exposures and a health outcome. The current stepwise selection technique ignores stochastic errors and suffers from a lack of stability. The alternative LASSO-penalized regression model can be applied to detect significant predictors from a pool of candidate variables. However, this technique is prone to false positives and tends to create excessive biases. It remains challenging to develop robust variable selection methods and enhance predictability.

MATERIAL AND METHODS

Two improved algorithms denoted the two-stage hybrid and bootstrap ranking procedures, both using a LASSO-type penalty, were developed for epidemiological association analysis. The performance of the proposed procedures and other methods including conventional LASSO, Bolasso, stepwise and stability selection models were evaluated using intensive simulation. In addition, methods were compared by using an empirical analysis based on large-scale survey data of hepatitis B infection-relevant factors among Guangdong residents.

RESULTS

The proposed procedures produced comparable or less biased selection results when compared to conventional variable selection models. In total, the two newly proposed procedures were stable with respect to various scenarios of simulation, demonstrating a higher power and a lower false positive rate during variable selection than the compared methods. In empirical analysis, the proposed procedures yielding a sparse set of hepatitis B infection-relevant factors gave the best predictive performance and showed that the procedures were able to select a more stringent set of factors. The individual history of hepatitis B vaccination, family and individual history of hepatitis B infection were associated with hepatitis B infection in the studied residents according to the proposed procedures.

CONCLUSIONS

The newly proposed procedures improve the identification of significant variables and enable us to derive a new insight into epidemiological association analysis.

Constraint programming based biomarker optimization

Efficient and intuitive characterization of biological big data is becoming a major challenge for modern bio-OMIC based scientists. Interactive visualization and exploration of big data is proven to be one of the successful solutions. Most of the existing feature selection algorithms do not allow the interactive inputs from users in the optimizing process of feature selection. This study investigates this question as fixing a few user-input features in the finally selected feature subset and formulates these user-input features as constraints for a programming model. The proposed algorithm, fsCoP (feature selection based on constrained programming), performs well similar to or much better than the existing feature selection algorithms, even with the constraints from both literature and the existing algorithms. An fsCoP biomarker may be intriguing for further wet lab validation, since it satisfies both the classification optimization function and the biomedical knowledge. fsCoP may also be used for the interactive exploration of bio-OMIC big data by interactively adding user-defined constraints for modeling.

AP-57/C10orf99 is a new type of multifunctional antimicrobial peptide

Antimicrobial peptides (AMPs) are an evolutionarily conserved component of the innate immune response that provides host defence at skin and mucosal surfaces. Here, we report the identification and characterization of a new type human AMPs, termed AP-57 (Antimicrobial Peptide with 57 amino acid residues), which is also known as C10orf99 (chromosome 10 open reading frame 99). AP-57 is a short basic amphiphilic peptide with four cysteines and a net charge +14 (MW = 6.52, PI = 11.28). The highest expression of AP-57 were detected in the mucosa of stomach and colon through immunohistochemical assay. Epithelium of skin and esophagus show obvious positive staining and strong positive staining were also observed in some tumor and/or their adjacent tissues, such as esophagus cancer, hepatocellular carcinoma, squamous cell carcinoma and invasive ductal carcinoma. AP-57 exhibited broad-spectrum antimicrobial activities against Gram-positive Staphylococcus aureus, Actinomyce, and Fungi Aspergillus niger as well as mycoplasma and lentivirus. AP-57 also exhibited DNA binding capacity and specific cytotoxic effects against human B-cell lymphoma Raji. Compared with other human AMPs, AP-57 has its distinct characteristics, including longer sequence length, four cysteines, highly cationic character, cell-specific toxicity, DNA binding and tissue-specific expressing patterns. Together, AP-57 is a new type of multifunctional AMPs worthy further investigation.

CSBF/C10orf99, a novel potential cytokine, inhibits colon cancer cell growth through inducing G1 arrest

Cytokines are soluble proteins that exert their functions by binding specific receptors. Many cytokines play essential roles in carcinogenesis and have been developed for the treatment of cancer. In this study, we identified a novel potential cytokine using immunogenomics designated colon-derived SUSD2 binding factor (CSBF), also known as chromosome 10 open reading frame 99 (C10orf99). CSBF/C10orf99 is a classical secreted protein with predicted molecular mass of 6.5 kDa, and a functional ligand of Sushi Domain Containing 2 (SUSD2). CSBF/C10orf99 has the highest expression level in colon tissue. Both CSBF/C10orf99 and SUSD2 are down-regulated in colon cancer tissues and cell lines with different regulation mechanisms. CSBF/C10orf99 interacts with SUSD2 to inhibit colon cancer cell growth and induce G1 cell cycle arrest by down-regulating cyclin D and cyclin-dependent kinase 6 (CDK6). CSBF/C10orf99 displays a bell-shaped activity curve with the optimal effect at ~10 ng/ml. Its growth inhibitory effects can be blocked by sSUSD2-Fc soluble protein. Our results suggest that CSBF/C10orf99 is a novel potential cytokine with tumor suppressor functions.

Integrative clinical transcriptomics analyses for new therapeutic intervention strategies: a psoriasis case study

Psoriasis is a chronic inflammatory skin disease with complex pathological features and unmet pharmacotherapy needs. Here, we present a framework for developing new therapeutic intervention strategies for psoriasis by utilizing publicly available clinical transcriptomics data sets. By exploring the underlying molecular mechanisms of psoriasis, the effects of subsequent perturbation of these mechanisms by drugs and an integrative analysis, we propose a psoriasis disease signature, identify potential drug repurposing opportunities and present novel target selection methodologies. We anticipate that the outlined methodology or similar approaches will further support biomarker discovery and the development of new drugs for psoriasis.