The impact of ordinal scales on Gaussian mixture recovery.

Gaussian mixture models (GMMs) are a popular and versatile tool for exploring heterogeneity in multivariate continuous data. Arguably the most popular way to estimate GMMs is via the expectation-maximization (EM) algorithm combined with model selection using the Bayesian information criterion (BIC). If the GMM is correctly specified, this estimation procedure has been demonstrated to have high recovery performance. However, in many situations, the data are not continuous but ordinal, for example when assessing symptom severity in medical data or modeling the responses in a survey. For such situations, it is unknown how well the EM algorithm and the BIC perform in GMM recovery. In the present paper, we investigate this question by simulating data from various GMMs, thresholding them in ordinal categories and evaluating recovery performance. We show that the number of components can be estimated reliably if the number of ordinal categories and the number of variables is high enough. However, the estimates of the parameters of the component models are biased independent of sample size. Finally, we discuss alternative modeling approaches which might be adopted for the situations in which estimating a GMM is not acceptable.

Evolution of DNA Methylation Across Ecdysozoa.

DNA methylation is a crucial, abundant mechanism of gene regulation in vertebrates. It is less prevalent in many other metazoan organisms and completely absent in some key model species, such as Drosophila melanogaster and Caenorhabditis elegans. We report here a comprehensive study of the presence and absence of DNA methyltransferases (DNMTs) in 138 Ecdysozoa, covering Arthropoda, Nematoda, Priapulida, Onychophora, and Tardigrada. Three of these phyla have not been investigated for the presence of DNA methylation before. We observe that the loss of individual DNMTs independently occurred multiple times across ecdysozoan phyla. We computationally predict the presence of DNA methylation based on CpG rates in coding sequences using an implementation of Gaussian Mixture Modeling, MethMod. Integrating both analysis we predict two previously unknown losses of DNA methylation in Ecdysozoa, one within Chelicerata (Mesostigmata) and one in Tardigrada. In the early-branching Ecdysozoa Priapulus caudatus, we predict the presence of a full set of DNMTs and the presence of DNA methylation. We are therefore showing a very diverse and independent evolution of DNA methylation in different ecdysozoan phyla spanning a phylogenetic range of more than 700 million years.

A novel meta-analytic approach: mining frequent co-activation patterns in neuroimaging databases.

In recent years, coordinate-based meta-analyses have become a powerful and widely used tool to study co-activity across neuroimaging experiments, a development that was supported by the emergence of large-scale neuroimaging databases like BrainMap. However, the evaluation of co-activation patterns is constrained by the fact that previous coordinate-based meta-analysis techniques like Activation Likelihood Estimation (ALE) and Multilevel Kernel Density Analysis (MKDA) reveal all brain regions that show convergent activity within a dataset without taking into account actual within-experiment co-occurrence patterns. To overcome this issue we here propose a novel meta-analytic approach named PaMiNI that utilizes a combination of two well-established data-mining techniques, Gaussian mixture modeling and the Apriori algorithm. By this, PaMiNI enables a data-driven detection of frequent co-activation patterns within neuroimaging datasets. The feasibility of the method is demonstrated by means of several analyses on simulated data as well as a real application. The analyses of the simulated data show that PaMiNI identifies the brain regions underlying the simulated activation foci and perfectly separates the co-activation patterns of the experiments in the simulations. Furthermore, PaMiNI still yields good results when activation foci of distinct brain regions become closer together or if they are non-Gaussian distributed. For the further evaluation, a real dataset on working memory experiments is used, which was previously examined in an ALE meta-analysis and hence allows a cross-validation of both methods. In this latter analysis, PaMiNI revealed a fronto-parietal "core" network of working memory and furthermore indicates a left-lateralization in this network. Finally, to encourage a widespread usage of this new method, the PaMiNI approach was implemented into a publicly available software system.

Interpretation of QSAR Models: Mining Structural Patterns Taking into Account Molecular Context.

The study focused on QSAR model interpretation. The goal was to develop a workflow for the identification of molecular fragments in different contexts important for the property modelled. Using a previously established approach - Structural and physicochemical interpretation of QSAR models (SPCI) - fragment contributions were calculated and their relative influence on the compounds' properties characterised. Analysis of the distributions of these contributions using Gaussian mixture modelling was performed to identify groups of compounds (clusters) comprising the same fragment, where these fragments had substantially different contributions to the property studied. SMARTSminer was used to detect patterns discriminating groups of compounds from each other and visual inspection if the former did not help. The approach was applied to analyse the toxicity, in terms of 40 hour inhibition of growth, of 1984 compounds to Tetrahymena pyriformis. The results showed that the clustering technique correctly identified known toxicophoric patterns: it detected groups of compounds where fragments have specific molecular context making them contribute substantially more to toxicity. The results show the applicability of the interpretation of QSAR models to retrieve reasonable patterns, even from data sets consisting of compounds having different mechanisms of action, something which is difficult to achieve using conventional pattern/data mining approaches.

Region-Specific Methylation Profiling in Acute Myeloid Leukemia.

Alteration of DNA methylation level in cancer diseases leads to deregulation of gene expression-silencing of tumor suppressor genes and enhancing of protooncogenes. There are several tools devoted to the problem of identification of CpG sites' demethylation but majority of them focuses on single site level and does not allow for quantification of region methylation changes. The aim was to create an adaptive algorithm supporting detection of differentially methylated CpG sites and genomic regions specific for acute myeloid leukemia. Knowledge on AML methylation fingerprint helps in better understanding the epigenetics of leukemogenesis. Proposed algorithm is data driven and does not use predefined quantification thresholds. Gaussian mixture modeling supports classification of CpG sites to several levels of demethylation. p value integration allows for translation from single site demethylation to the demethylation of gene promoter and body regions. Methylation profiles of healthy controls and AML patients were examined (GEO:GSE63409). The differences in whole genome methylation profiles were observed. The methylation profile differs significantly among genomic regions. The lowest methylation level was observed for promoter regions, while sites from intergenic regions were by average higher methylated. The observed number of AML related down methylated sites has not substantially exceeded the expected number by chance. Intergenic regions were characterized by the highest percentage of AML up methylated sites. Methylation enhancement/diminution is the most frequent for intergenic region while methylation compensation (positive or negative) is specific for promoter regions. Functional analysis performed for AML down methylated or extreme high up methylated genes showed strong connection to the leukemic processes.

Diagnostic performance of Elecsys immunoassays for cerebrospinal fluid Alzheimer's disease biomarkers in a nonacademic, multicenter memory clinic cohort: The ABIDE project.

INTRODUCTION: We compared the automated Elecsys and manual Innotest immunoassays for cerebrospinal fluid (CSF) Alzheimer's disease biomarkers in a multicenter diagnostic setting. METHODS: We collected CSF samples from 137 participants in eight local memory clinics. Amyloid ß(1-42) (Aß42), total tau (t-tau), and phosphorylated tau (p-tau) were centrally analyzed with Innotest and Elecsys assays. Concordances between methods were assessed. RESULTS: Biomarker results strongly correlated between assays with Spearman's ρ 0.94 for Aß42, 0.98 for t-tau, and 0.98 for p-tau. Using Gaussian mixture modeling, cohort-specific cut-points were estimated at 1092 pg/mL for Aß42, 235 pg/mL for t-tau, and 24 pg/mL for p-tau. We found an excellent concordance of biomarker abnormality between assays of 97% for Aß42 and 96% for both t-tau and p-tau. DISCUSSION: The high concordances between Elecsys and Innotest in this nonacademic, multicenter cohort support the use of Elecsys for CSF Alzheimer's disease diagnostics and allow conversion of results between methods.

A Novel Unsupervised Segmentation Approach Quantifies Tumor Tissue Populations Using Multiparametric MRI: First Results with Histological Validation.

PURPOSE: We aimed to precisely estimate intra-tumoral heterogeneity using spatially regularized spectral clustering (SRSC) on multiparametric MRI data and compare the efficacy of SRSC with the previously reported segmentation techniques in MRI studies. PROCEDURES: Six NMRI nu/nu mice bearing subcutaneous human glioblastoma U87 MG tumors were scanned using a dedicated small animal 7T magnetic resonance imaging (MRI) scanner. The data consisted of T2 weighted images, apparent diffusion coefficient maps, and pre- and post-contrast T2 and T2* maps. Following each scan, the tumors were excised into 2-3-mm thin slices parallel to the axial field of view and processed for histological staining. The MRI data were segmented using SRSC, K-means, fuzzy C-means, and Gaussian mixture modeling to estimate the fractional population of necrotic, peri-necrotic, and viable regions and validated with the fractional population obtained from histology. RESULTS: While the aforementioned methods overestimated peri-necrotic and underestimated viable fractions, SRSC accurately predicted the fractional population of all three tumor tissue types and exhibited strong correlations (rnecrotic = 0.92, rperi-necrotic = 0.82 and rviable = 0.98) with the histology. CONCLUSIONS: The precise identification of necrotic, peri-necrotic and viable areas using SRSC may greatly assist in cancer treatment planning and add a new dimension to MRI-guided tumor biopsy procedures.

Three-dimensional segmentation of breast masses from digital breast tomosynthesis images.

Assessment of three-dimensional (3-D) morphology and volume of breast masses is important for cancer diagnosis, staging, and treatment but cannot be derived from conventional mammography. Digital breast tomosynthesis (DBT) provides data from which 3-D mass segmentation could be obtained. Our method combined Gaussian mixture models based on intensity and a texture measure indicative of in-focus structure, gray-level variance. Thresholding these voxel probabilities, weighted by distance to the estimated mass center, gave the final 3-D segmentation. Evaluation used 40 masses annotated twice by a consultant radiologist on in-focus slices in two diagnostic views. Human intraobserver variability was assessed as the overlap between repeated annotations (median 77% and range 25% to 91%). Comparing the segmented mass outline with probability-weighted ground truth from these annotations, median agreement was 68%, and range was 7% to 88%. Annotated and segmented diameters correlated well with histological mass size (both Spearman's rank correlations [Formula: see text]). The volumetric segmentation demonstrated better agreement with tumor volumes estimated from pathology than volume derived from radiological annotations (95% limits of agreement [Formula: see text] to 11 ml and [Formula: see text] to 41 ml, respectively). We conclude that it is feasible to assess 3-D mass morphology and volume from DBT, and the method has the potential to aid breast cancer management.

Methadone induces hypermethylation of human DNA.

BACKGROUND: Increased global DNA methylation in the blood of patients chronically exposed to opioids had been interpreted as an indication of an epigenetic action of this drug class. MATERIALS & METHODS: To strengthen the causality, human MCF7 cells were cultured in media with the addition of several known or potential modulators of DNA methylation including methadone. RESULTS: Following 3 days of incubation with several different known or potential epigenetic modulators, global DNA methylation, quantified at LINE-1 CpG islands, showed a large variability across all treatments ranging from 27.8 to 63%. Based on distribution analysis of the global methylation of human DNA exposed to various potential modulators, present in vitro experiments showed that treatment with the opioid methadone was associated with an increased probability of hypermethylation. CONCLUSION: This strengthens the evidence that opioids interfere with mechanisms of classical epigenetics.

Automatic segmentation of corpus callosum using Gaussian mixture modeling and Fuzzy C means methods.

This paper presents a comparative study of the success and performance of the Gaussian mixture modeling and Fuzzy C means methods to determine the volume and cross-sectionals areas of the corpus callosum (CC) using simulated and real MR brain images. The Gaussian mixture model (GMM) utilizes weighted sum of Gaussian distributions by applying statistical decision procedures to define image classes. In the Fuzzy C means (FCM), the image classes are represented by certain membership function according to fuzziness information expressing the distance from the cluster centers. In this study, automatic segmentation for midsagittal section of the CC was achieved from simulated and real brain images. The volume of CC was obtained using sagittal sections areas. To compare the success of the methods, segmentation accuracy, Jaccard similarity and time consuming for segmentation were calculated. The results show that the GMM method resulted by a small margin in more accurate segmentation (midsagittal section segmentation accuracy 98.3% and 97.01% for GMM and FCM); however the FCM method resulted in faster segmentation than GMM. With this study, an accurate and automatic segmentation system that allows opportunity for quantitative comparison to doctors in the planning of treatment and the diagnosis of diseases affecting the size of the CC was developed. This study can be adapted to perform segmentation on other regions of the brain, thus, it can be operated as practical use in the clinic.