MCMC implementation of the optimal Bayesian classifier for non-Gaussian models: model-based RNA-Seq classification

Background

Sequencing datasets consist of a finite number of reads which map to specific regions of a reference genome. Most effort in modeling these datasets focuses on the detection of univariate differentially expressed genes. However, for classification, we must consider multiple genes and their interactions.

Results

Thus, we introduce a hierarchical multivariate Poisson model (MP) and the associated optimal Bayesian classifier (OBC) for classifying samples using sequencing data. Lacking closed-form solutions, we employ a Monte Carlo Markov Chain (MCMC) approach to perform classification. We demonstrate superior or equivalent classification performance compared to typical classifiers for two synthetic datasets and over a range of classification problem difficulties. We also introduce the Bayesian minimum mean squared error (MMSE) conditional error estimator and demonstrate its computation over the feature space. In addition, we demonstrate superior or leading class performance over an RNA-Seq dataset containing two lung cancer tumor types from The Cancer Genome Atlas (TCGA).

Conclusions

Through model-based, optimal Bayesian classification, we demonstrate superior classification performance for both synthetic and real RNA-Seq datasets. A tutorial video and Python source code is available under an open source license at http://bit.ly/1gimnss.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-014-0401-3) contains supplementary material, which is available to authorized users.     

Inferring copy number and genotype in tumour exome data

Background

Using whole exome sequencing to predict aberrations in tumours is a cost effective alternative to whole genome sequencing, however is predominantly used for variant detection and infrequently utilised for detection of somatic copy number variation.

Results

We propose a new method to infer copy number and genotypes using whole exome data from paired tumour/normal samples. Our algorithm uses two Hidden Markov Models to predict copy number and genotypes and computationally resolves polyploidy/aneuploidy, normal cell contamination and signal baseline shift. Our method makes explicit detection on chromosome arm level events, which are commonly found in tumour samples. The methods are combined into a package named ADTEx (Aberration Detection in Tumour Exome). We applied our algorithm to a cohort of 17 in-house generated and 18 TCGA paired ovarian cancer/normal exomes and evaluated the performance by comparing against the copy number variations and genotypes predicted using Affymetrix SNP 6.0 data of the same samples. Further, we carried out a comparison study to show that ADTEx outperformed its competitors in terms of precision and F-measure.

Conclusions

Our proposed method, ADTEx, uses both depth of coverage ratios and B allele frequencies calculated from whole exome sequencing data, to predict copy number variations along with their genotypes. ADTEx is implemented as a user friendly software package using Python and R statistical language. Source code and sample data are freely available under GNU license (GPLv3) at http://adtex.sourceforge.net/.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-732) contains supplementary material, which is available to authorized users.     

SHEAR: sample heterogeneity estimation and assembly by reference

Background

Personal genome assembly is a critical process when studying tumor genomes and other highly divergent sequences. The accuracy of downstream analyses, such as RNA-seq and ChIP-seq, can be greatly enhanced by using personal genomic sequences rather than standard references. Unfortunately, reads sequenced from these types of samples often have a heterogeneous mix of various subpopulations with different variants, making assembly extremely difficult using existing assembly tools. To address these challenges, we developed SHEAR (Sample Heterogeneity Estimation and Assembly by Reference; http://vk.cs.umn.edu/SHEAR), a tool that predicts SVs, accounts for heterogeneous variants by estimating their representative percentages, and generates personal genomic sequences to be used for downstream analysis.

Results

By making use of structural variant detection algorithms, SHEAR offers improved performance in the form of a stronger ability to handle difficult structural variant types and better computational efficiency. We compare against the lead competing approach using a variety of simulated scenarios as well as real tumor cell line data with known heterogeneous variants. SHEAR is shown to successfully estimate heterogeneity percentages in both cases, and demonstrates an improved efficiency and better ability to handle tandem duplications.

Conclusion

SHEAR allows for accurate and efficient SV detection and personal genomic sequence generation. It is also able to account for heterogeneous sequencing samples, such as from tumor tissue, by estimating the subpopulation percentage for each heterogeneous variant.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-84) contains supplementary material, which is available to authorized users.     

Oral Doxycycline Reduces Pterygium Lesions; Results from a Double Blind,Randomized, Placebo Controlled Clinical Trial

Purpose

To determine whether oral doxycycline treatment reduces pterygium lesions.

Design

Double blind, randomized, placebo controlled clinical trial.

Participants

98 adult patients with primary pterygium.

Methods

Patients were randomly assigned to receive 100 mg oral doxycycline twice a day (49 subjects), or placebo (49 subjects), for 30 days. Photographs of the lesion were taken at the time of recruitment and at the end of the treatment. Follow-up sessions were performed 6 and 12 months post-treatment. Statistical analyses for both continuous and categorical variables were applied. p values of less than 0.05 were considered to indicate statistical significance.

Main Outcome Measures

The primary endpoint was the change in lesion size after 30 days of treatment.

Results

The primary endpoint was not met for the whole population but subgroup analysis showed that doxycycline was effective in patients of Caucasian origin while other ethnicities, mostly Hispanic, did not respond to the treatment. Moreover, there was a correlation between age and better response (p = 0.003). Adverse events were uncommon, mild, and in agreement with previous reports on short doxycycline treatments.

Conclusions

Oral doxycycline was superior to placebo for the treatment of primary pterygia in older Caucasian patients. These findings support the use of doxycycline for pterygium treatment in particular populations.

Trial Registration

European Union Clinical Trials Register EudraCT 2008-007178-39     

Assessment of Still and Moving Images in the Diagnosis of Gastric Lesions Using Magnifying Narrow-Band Imaging in a Prospective Multicenter Trial

Objectives

Magnifying narrow-band imaging (M-NBI) is more accurate than white-light imaging for diagnosing small gastric cancers. However, it is uncertain whether moving M-NBI images have additional effects in the diagnosis of gastric cancers compared with still images.

Design

A prospective multicenter cohort study.

Methods

To identify the additional benefits of moving M-NBI images by comparing the diagnostic accuracy of still images only with that of both still and moving images. Still and moving M-NBI images of 40 gastric lesions were obtained by an expert endoscopist prior to this prospective multicenter cohort study. Thirty-four endoscopists from ten different Japanese institutions participated in the prospective multicenter cohort study. Each study participant was first tested using only still M-NBI images (still image test), then tested 1 month later using both still and moving M-NBI images (moving image test). The main outcome was a difference in the diagnostic accuracy of cancerous versus noncancerous lesions between the still image test and the moving image test.

Results

Thirty-four endoscopists were analysed. There were no significant difference of cancerous versus noncancerous lesions between still and moving image tests in the diagnostic accuracy (59.9% versus 61.5%), sensitivity (53.4% versus 55.9%), and specificity (67.0% versus 67.6%). And there were no significant difference in the diagnostic accuracy between still and moving image tests of demarcation line (65.4% versus 65.5%), microvascular pattern (56.7% versus 56.9%), and microsurface pattern (48.1% versus 50.9%). Diagnostic accuracy showed no significant difference between the still and moving image tests in the subgroups of endoscopic findings of the lesions.

Conclusions

The addition of moving M-NBI images to still M-NBI images does not improve the diagnostic accuracy for gastric lesions. It is reasonable to concentrate on taking sharp still M-NBI images during endoscopic observation and use them for diagnosis.

Trial registration

Umin.ac.jp UMIN-CTR000008048     

LDL-Migration Index (LDL-MI), an Indicator of Small Dense Low-Density Lipoprotein (sdLDL), Is Higher in Non-Alcoholic Steatohepatitis than in Non-Alcoholic Fatty Liver: A Multicenter Cross-Sectional Study

Background

Non-alcoholic fatty liver disease (NAFLD) is associated with increased risks of atherosclerotic diseases, including cardiovascular disease. However, the difference in risk between patients with non-alcoholic fatty liver (NAFL) and non-alcoholic steatohepatitis (NASH) has not yet been determined. Accumulating evidence has shown that high amounts of small dense low-density lipoprotein (sdLDL) are closely associated with atherosclerotic diseases. This study investigated differences in risk factors for atherosclerotic diseases, especially LDL-migration index (LDL-MI), an indicator of sdLDL, between patients with NAFL and NASH.

Methods

LDL-MI was analyzed in a primary cohort of 156 patients with NAFLD, including 53 with NAFL and 103 with NASH, and a validation cohort of 69 patients with NAFLD, including 25 with NAFL and 44 with NASH.

Results

In the primary cohort, NASH was associated with elevated LDL-MI (p = 0.039). Multiple regression analysis showed that NASH and the non-use of lipid lowering medications were independently correlated with higher LDL-MI in all patients with NAFLD. Among patients not on lipid lowering medications, those with NASH had significantly higher LDL-MI than those with NAFL (p = 0.001). These findings were confirmed in a validation cohort, in that LDL-MI was significantly higher in patients with NASH than with NAFL (p = 0.043).

Conclusion

This study is the first to show that LDL-MI, an indicator of sdLDL, was higher in patients with NASH than with NAFL, suggesting that the risk of atherosclerotic diseases may be higher in NASH than NAFL. Patients with NASH should be followed closely, especially for the progression of liver pathology and atherosclerotic diseases.

Trial Registration

UMIN000009614     

Rituximab and abatacept but not tocilizumab impair antibody response to pneumococcal conjugate vaccine in patients with rheumatoid arthritis

Introduction

The objective of the study was to investigate the impact of newer biologic treatments including rituximab, abatacept and tocilizumab on antibody response following pneumococcal vaccination using a 7-valent conjugate vaccine in patients with established rheumatoid arthritis (RA).

Methods

Patients with RA receiving rituximab, abatacept or tocilizumab as monotherapy or combined with methotrexate (MTX) participated in the study. Specific IgG antibodies against 23F and 6B serotypes were measured at vaccination and 4 to 6 weeks after vaccination using standardised ELISA. Geometric mean antibody levels (GML) were calculated. Antibody response (AR) was defined as the ratio between post- and pre-vaccination antibody levels and a positive antibody response (posAR) was AR ≥2.

Results

In total, 88 patients were enrolled in the study. Of 55 patients treated with rituximab, 26 (46%) were on concomitant MTX. Of patients receiving abatacept (n = 17) and tocilizumab (n = 16) biologic treatment was given in combination with MTX in 13 (76%) and 9 (56%) patients, respectively. Patients treated with rituximab had significantly lower AR compared to those on tocilizumab, as well as compared to previously reported RA patients on MTX and controls (spondylarthropathy patients treated with NSAIDs and/or analgesics). In total, 10.3% of patients on rituximab monotherapy and no patient on rituximab + MTX had posAR for both serotypes. For abatacept and tocilizumab the corresponding figures were 17.6% and 50%.

Conclusion

In this cohort of patients with established RA, treatment with rituximab and abatacept was associated with diminished antibody response but this was most pronounced for rituximab. Pneumococcal conjugate vaccine administrated during ongoing tocilizumab treatment seems to be associated with sufficient antibody response. Pneumococcal vaccination should preferably be encouraged before initiation of rituximab or abatacept treatment.

Trial registration

NCT00828997 and EudraCT EU 2007-006539-29.     

Improving Adherence and Clinical Outcomes in Self-Guided Internet Treatment for Anxiety and Depression: Randomised Controlled Trial

Background

Depression and anxiety are common, disabling and chronic. Self-guided internet-delivered treatments are popular, but few people complete them. New strategies are required to realise their potential.

Aims

To evaluate the effect of automated emails on the effectiveness, safety, and acceptability of a new automated transdiagnostic self-guided internet-delivered treatment, the Wellbeing Course, for people with depression and anxiety.

Method

A randomised controlled trial was conducted through the website: www.ecentreclinic.org. Two hundred and fifty seven people with elevated symptoms were randomly allocated to the 8 week course either with or without automated emails, or to a waitlist control group. Primary outcome measures were the Patient Health Questionnaire 9-Item (PHQ-9) and the Generalized Anxiety Disorder 7-Item (GAD-7).

Results

Participants in the treatment groups had lower PHQ-9 and GAD-7 scores at post-treatment than controls. Automated emails increased rates of course completion (58% vs. 35%), and improved outcomes in a subsample with elevated symptoms.

Conclusions

The new self-guided course was beneficial, and automated emails facilitated outcomes. Further attention to strategies that facilitate adherence, learning, and safety will help realise the potential of self-guided interventions.

Trial Registration

Australian and New Zealand Clinical Trials Registry ACTRN12610001058066     

Effect of High- versus Low-Intensity Supervised Aerobic and Resistance Training on Modifiable Cardiovascular Risk Factors in Type 2 Diabetes; The Italian Diabetes and Exercise Study (IDES)

Background

While current recommendations on exercise type and volume have strong experimental bases, there is no clear evidence from large-sized studies indicating whether increasing training intensity provides additional benefits to subjects with type 2 diabetes.

Objective

To compare the effects of moderate-to-high intensity (HI) versus low-to-moderate intensity (LI) training of equal energy cost, i.e. exercise volume, on modifiable cardiovascular risk factors.

Design

Pre-specified sub-analysis of the Italian Diabetes and Exercise Study (IDES), a randomized multicenter prospective trial comparing a supervised exercise intervention with standard care for 12 months (2005–2006).

Setting

Twenty-two outpatient diabetes clinics across Italy.

Patients

Sedentary patients with type 2 diabetes assigned to twice-a-week supervised progressive aerobic and resistance training plus exercise counseling (n = 303).

Interventions

Subjects were randomized by center to LI (n = 142, 136 completed) or HI (n = 161, 152 completed) progressive aerobic and resistance training, i.e. at 55% or 70% of predicted maximal oxygen consumption and at 60% or 80% of predicted 1-Repetition Maximum, respectively, of equal volume.

Main Outcome Measure(s)

Hemoglobin (Hb) A_1c and other cardiovascular risk factors; 10-year coronary heart disease (CHD) risk scores.

Results

Volume of physical activity, both supervised and non-supervised, was similar in LI and HI participants. Compared with LI training, HI training produced only clinically marginal, though statistically significant, improvements in HbA_1c (mean difference −0.17% [95% confidence interval −0.44,0.10], P = 0.03), triglycerides (−0.12 mmol/l [−0.34,0.10], P = 0.02) and total cholesterol (−0.24 mmol/l [−0.46, −0.01], P = 0.04), but not in other risk factors and CHD risk scores. However, intensity was not an independent predictor of reduction of any of these parameters. Adverse event rate was similar in HI and LI subjects.

Conclusions

Data from the large IDES cohort indicate that, in low-fitness individuals such as sedentary subjects with type 2 diabetes, increasing exercise intensity is not harmful, but does not provide additional benefits on cardiovascular risk factors.

Trial Registration

www.ISRCTN.org ISRCTN-04252749.     

Correlating changes in lung function with patient outcomes in chronic obstructive pulmonary disease: a pooled analysis

Background

Relationships between improvements in lung function and other clinical outcomes in chronic obstructive pulmonary disease (COPD) are not documented extensively. We examined whether changes in trough forced expiratory volume in 1 second (FEV₁) are correlated with changes in patient-reported outcomes.

Methods

Pooled data from three indacaterol studies (n = 3313) were analysed. Means and responder rates for outcomes including change from baseline in Transition Dyspnoea Index (TDI), St. George''s Respiratory Questionnaire (SGRQ) scores (at 12, 26 and 52 weeks), and COPD exacerbation frequency (rate/year) were tabulated across categories of ΔFEV₁. Also, generalised linear modelling was performed adjusting for covariates such as baseline severity and inhaled corticosteroid use.

Results

With increasing positive ΔFEV₁, TDI and ΔSGRQ improved at all timepoints, exacerbation rate over the study duration declined (P < 0.001). Individual-level correlations were 0.03-0.18, but cohort-level correlations were 0.79-0.95. At 26 weeks, a 100 ml increase in FEV₁ was associated with improved TDI (0.46 units), ΔSGRQ (1.3-1.9 points) and exacerbation rate (12% decrease). Overall, adjustments for baseline covariates had little impact on the relationship between ΔFEV₁ and outcomes.

Conclusions

These results suggest that larger improvements in FEV₁ are likely to be associated with larger patient-reported benefits across a range of clinical outcomes.

Trial Registration

ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT00393458","term_id":"NCT00393458"}}NCT00393458, {"type":"clinical-trial","attrs":{"text":"NCT00463567","term_id":"NCT00463567"}}NCT00463567, and {"type":"clinical-trial","attrs":{"text":"NCT00624286","term_id":"NCT00624286"}}NCT00624286     

BACA: bubble chArt to compare annotations

Background

DAVID is the most popular tool for interpreting large lists of gene/proteins classically produced in high-throughput experiments. However, the use of DAVID website becomes difficult when analyzing multiple gene lists, for it does not provide an adequate visualization tool to show/compare multiple enrichment results in a concise and informative manner.

Result

We implemented a new R-based graphical tool, BACA (Bubble chArt to Compare Annotations), which uses the DAVID web service for cross-comparing enrichment analysis results derived from multiple large gene lists. BACA is implemented in R and is freely available at the CRAN repository (http://cran.r-project.org/web/packages/BACA/).

Conclusion

The package BACA allows R users to combine multiple annotation charts into one output graph by passing DAVID website.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0477-4) contains supplementary material, which is available to authorized users.     

PMS: A Panoptic Motif Search Tool

Background

Identification of DNA/Protein motifs is a crucial problem for biologists. Computational techniques could be of great help in this identification. In this direction, many computational models for motifs have been proposed in the literature.

Methods

One such important model is the motif model. In this paper we describe a motif search web tool that predominantly employs this motif model. This web tool exploits the state-of-the art algorithms for solving the motif search problem.

Results

The online tool has been helping scientists identify many unknown motifs. Many of our predictions have been successfully verified as well. We hope that this paper will expose this crucial tool to many more scientists.

Availability and requirements

Project name: PMS - Panoptic Motif Search Tool. Project home page: http://pms.engr.uconn.edu or http://motifsearch.com. Licence: PMS tools will be readily available to any scientist wishing to use it for non-commercial purposes, without restrictions. The online tool is freely available without login.     

The Effectiveness of an Online Support Group for Members of the Community with Depression: A Randomised Controlled Trial

Background

Internet support groups (ISGs) are popular, particularly among people with depression, but there is little high quality evidence concerning their effectiveness.

Aim

The study aimed to evaluate the efficacy of an ISG for reducing depressive symptoms among community members when used alone and in combination with an automated Internet-based psychotherapy training program.

Method

Volunteers with elevated psychological distress were identified using a community-based screening postal survey. Participants were randomised to one of four 12-week conditions: depression Internet Support Group (ISG), automated depression Internet Training Program (ITP), combination of the two (ITP+ISG), or a control website with delayed access to e-couch at 6 months. Assessments were conducted at baseline, post-intervention, 6 and 12 months.

Results

There was no change in depressive symptoms relative to control after 3 months of exposure to the ISG. However, both the ISG alone and the combined ISG+ITP group showed significantly greater reduction in depressive symptoms at 6 and 12 months follow-up than the control group. The ITP program was effective relative to control at post-intervention but not at 6 months.

Conclusions

ISGs for depression are promising and warrant further empirical investigation.

Trial Registration

Controlled-Trials.com ISRCTN65657330     

Predictive Factors for Sustained Virological Response after Treatment with Pegylated Interferon α-2a and Ribavirin in Patients Infected with HCV Genotypes 2 and 3

Background

Previous trials have often defined genotype 2 and 3 patients as an “easy to treat” group and guidelines recommend similar management.

Aims

The present study looks for differences between the two genotypes and analyzes predictive factors for SVR.

Methods

Prospective, community-based cohort study involving 421 physicians throughout Germany. The analysis includes 2,347 patients with untreated chronic HCV genotype 2 (n = 391) and 3 (n = 1,956) infection treated with PEG-IFN α-2a plus ribavirin between August 2007 and July 2012.

Results

When compared with genotype 2 patients, those with genotype 3 were younger, had a shorter duration of infection, lower values of total cholesterol, LDL cholesterol and BMI, a higher frequency of drug use as infection mode and male gender (p<0.0001, respectively), and a higher APRI score (p<0.005). SVR was higher in genotype 2 when compared with genotype 3 (64.7% vs. 56.9%, p = 0.004). By multivariate analysis of genotype 2 patients, low baseline γ -GT and RVR predicted SVR. In genotype 3 age ≤45 years, cholesterol>130 mg/dl, a low APRI score, and a γ-GT ≥3-times ULN, RVR, and RBV starting dose were associated with SVR by multivariate analysis.

Conclusions

The present study corroborates that liver fibrosis is more pronounced in genotype 3 vs. 2. SVR is higher in genotype 2 versus genotype 3 partly because of follow-up problems in genotype 3 patients, in particular in those infected by drug use. Thus, subgroups of genotype 3 patients have adherence problems and need special attention also because they often have significant liver fibrosis.

Trial Registration

Verband Forschender Arzneimittelhersteller e.V., Berlin, Germany ML21645 ClinicalTrials.gov {"type":"clinical-trial","attrs":{"text":"NCT02106156","term_id":"NCT02106156"}}NCT02106156     

COPD association and repeatability of blood biomarkers in the ECLIPSE cohort

Background

There is a need for biomarkers to better characterise individuals with COPD and to aid with the development of therapeutic interventions. A panel of putative blood biomarkers was assessed in a subgroup of the Evaluation of COPD Longitudinally to Identify Surrogate Endpoints (ECLIPSE) cohort.

Methods

Thirty-four blood biomarkers were assessed in 201 subjects with COPD, 37 ex-smoker controls with normal lung function and 37 healthy non-smokers selected from the ECLIPSE cohort. Biomarker repeatability was assessed using baseline and 3-month samples. Intergroup comparisons were made using analysis of variance, repeatability was assessed through Bland-Altman plots, and correlations between biomarkers and clinical characteristics were assessed using Spearman correlation coefficients.

Results

Fifteen biomarkers were significantly different in individuals with COPD when compared to former or non-smoker controls. Some biomarkers, including tumor necrosis factor-α and interferon-γ, were measurable in only a minority of subjects whilst others such as C-reactive protein showed wide variability over the 3-month replication period. Fibrinogen was the most repeatable biomarker and exhibited a weak correlation with 6-minute walk distance, exacerbation rate, BODE index and MRC dyspnoea score in COPD subjects. 33% (66/201) of the COPD subjects reported at least 1 exacerbation over the 3 month study with 18% (36/201) reporting the exacerbation within 30 days of the 3-month visit. CRP, fibrinogen interleukin-6 and surfactant protein-D were significantly elevated in those COPD subjects with exacerbations within 30 days of the 3-month visit compared with those individuals that did not exacerbate or whose exacerbations had resolved.

Conclusions

Only a few of the biomarkers assessed may be useful in diagnosis or management of COPD where the diagnosis is based on airflow obstruction (GOLD). Further analysis of more promising biomarkers may reveal utility in subsets of patients. Fibrinogen in particular has emerged as a potentially useful biomarker from this cohort and requires further investigation.

Trial Registration

SCO104960, clinicaltrials.gov identifier {"type":"clinical-trial","attrs":{"text":"NCT00292552","term_id":"NCT00292552"}}NCT00292552