1 - 36 of 36 results


Predicts histone modification and DNA methylation patterns from DNA motifs. Epigram is an analysis pipeline used to systematically identify DNA motifs that are predictive of epigenomic modifications. It reveals the cis-regulatory program that is read by the dynamic genetic network to shape the epigenome. In particular, Epigram’s motifs are significantly correlated with almost double the number of H3K27ac regions when compared to five times the number of known Transcription Factor (TF) motifs.


An algorithm implemented in R to identify disease specific hyper and hypomethylated genes. MethylMix is based on a beta mixture model to identify methylation states and compares them with the normal DNA methylation state. MethylMix introduces a novel metric, the "Differential Methylation value" or DM-value defined as the difference of a methylation state with the normal methylation state. Finally, matched gene expression data is used to identify, besides differential, transcriptionally predictive methylation states by focusing on methylation changes that effect gene expression.


Allows researchers and clinical diagnosticians to test single samples as well as large cohorts for subclass prediction of pediatric brain tumors. In the current version of MethPed the following groups are included; glioblastoma (GBM), pilocytic astrocytoma, medulloblastoma (Wnt, Shh, group 3 and group 4), diffuse intrinsic pontine glioma (DIPG), ependymoma and embryonal tumor with multilayered rosettes (ETMR). The MethPed R package can be used to efficiently classify pediatric brain tumors using DNA methylation profiles generated by the Illumina 450 K methylation arrays.


A computational strategy to predict DNA methylation by integrating cell-type specific 450K array data and common DNA sequence features. We developed a computational model that is trained on 14 tissues with both whole genome bisulfite sequencing and 450K array data. This model integrates information derived from the similarity of local methylation pattern between tissues, the methylation information of flanking CpG sites and the methylation tendency of flanking DNA sequences. When applying to a new sample, our model only requires input of 450K array data and avoids the need of histone modification or MeDIP-seq/MRE-seq data that are not always available, which significantly expands its applicability.

DIRECTION / Discriminative IntegRative whole Epigenome Classification at single nucleotide resoluTION

Allows to identify candidate genomic regions for differential hydroxymethylation as a first step in functional studies. DIRECTION is an in-silico, whole epigenome predictor of DNA methylation and 5-hmC status at single nucleotide resolution. It can be trained on shotgun sequencing-based mammalian methylation and hydroxymethylation datasets, by identifying and using available, correlated, high-throughput assays and genomic sequence-based traits as predictor variables.

H(O)TA / Hairpin (Oxidative) bisulfite sequencing Time course Analyzer

Infers (hydroxy-)methylation levels and efficiencies of the involved enzymes at a certain DNA locus. H(O)TA is based on the construction of two coupled Hidden Markov Models (HMMs). It is able to take into account all relevant conversion errors. This tool predicts only the methylation levels and efficiencies, merged with the corresponding unknown hydroxylation values, of a given region. It is capable of analyzing time course data from hairpin bisulfite sequencing and hairpin oxidative bisulfite sequencing in the same time.


Exploits the diagnostic potential of cell-free DNA by determining not only the presence but also the location of tumors. CancerLocator simultaneously infers the proportions and the tissue-of-origin of tumor-derived cell-free DNA in a blood sample using genome-wide DNA methylation data. It outperforms two established multi-class classification methods on simulations and real data, even with the low proportion of tumor-derived DNA in the cell-free DNA scenarios. This package also achieves promising results on patient plasma samples with low DNA methylation sequencing coverage.


star_border star_border star_border star_border star_border
star star star star star
Identifies DNA methylation sites via pseudo trinucleotide composition. We anticipate that the iDNA-Methyl predictor will become a useful high-throughput tool because (i) information of DNA methylation sites is important for both basic research and drug development and (ii) compared with the existing predictors in identifying the DNA methylation sites, it has remarkably higher success rates and a workable and publicly accessible web-server.


A web server for predicting the DNA methylation state of CpG dinucleotide using features inferred from three-dimensional genome topology (based on Hi-C) and DNA sequence patterns. DeepMethyl is based on a stacked denoising autoencoders deep learning algorithm. We used the experimental data from immortalised myelogenous leukemia (K562) and healthy lymphoblastoid (GM12878) cell lines to train the learning models and assess prediction performance. Using the methylation states of sequentially neighboring regions as one of the learning features, DeepMethyl achieved a blind test accuracy of 89.7% for GM12878 and 88.6% for K562. When the methylation states of sequentially neighboring regions are unknown, the accuracies are 84.82% for GM12878 and 72.01% for K562. We also analyzed the contribution of genome topological features inferred from Hi-C.

REMP / Repetitive Element Methylation Prediction

Predicts DNA methylation of locus-specific repetitive elements (RE) by learning surrounding genetic and epigenetic information. REMP is a machine learning-based tools to provide genome wide and single-base resolution of DNA methylation prediction on RE that are difficult to measure using array-based or sequencing-based platforms, which enables epigenome-wide association study (EWAS) and differentially methylated region (DMR) analysis on RE.

LYNE / Lineage Dynamics for Epigenetics

A Markov model framework to analyze DNA methylation dynamics across lineage specifications. LYNE takes precursor to descendant relationships into account and enables inference of CpG methylation dynamics. Using our model, we illustrate that (i) CpG site methylation status can be accurately reconstructed using data from related cell types at the same site, (ii) the single CpG site resolution of our methylation dynamics estimates enable the discovery of attributes, such as DNA sequence context, that correlate with CpG methylation dynamics and (iii) our models facilitate the identification of genomic regions with highly variable CpG methylation states that are likely functional.

CETS / Cell EpigenoType Specific

Is designed for the quantification and normalization of differing neuronal proportions in genome-scale DNA methylation datasets. The application of CETS quantification and transformation can reduce heterogeneity and improve replicability of epigenetic findings in the brain across cohorts. This tool will not only allow for the generation of novel data independent of cell heterogeneity based bias, but also allowing for a re-analysis of existing data sets. Application of CETS modeling to genome-wide DNA methylation data may lead to new level of understanding of epigenetic regulation in the brain and holds the potential to identify to novel discoveries related to the epigenetic basis of neurological and psychiatric phenotypes.


Implements a computational pattern recognition method that is used to predict the methylation landscape of human brain DNA. HDMFinder method can be applied both to CpG islands and to non-CpG island regions. It computes the methylation propensity for an 800-bp region centered on a CpG dinucleotide based on specific sequence features within the region. HDMFinder has a prediction accuracy of 86%, as validated with CpG regions for which methylation status has been experimentally determined. Using HDMFinder, the entire genomic methylation patterns for all 22 human autosomes have been depicted.


A support vector machine (SVM)-based method for the prediction of cytosine methylation in CpG dinucleotides. Initially a SVM module was developed from human data for the prediction of human-specific methylation sites. This module achieved a MCC and AUC of 0.501 and 0.814, respectively, when evaluated using a 5-fold cross-validation. The performance of this SVM-based module was better than the classifiers built using alternative machine learning and statistical algorithms including artificial neural networks, Bayesian statistics, and decision trees. Additional SVM modules were also developed based on mammalian- and vertebrate-specific methylation patterns. The SVM module based on human methylation patterns was used for genome-wide analysis of methylation sites. This analysis demonstrated that the percentage of methylated CpGs is higher in UTRs as compared to exonic and intronic regions of human genes.

CGPredictor / Cancer Grade Predictor

Localizes and examines biomarkers from strong self-similarity pattern on patients’ profiles. CGPredictor can facilitate the identification of distinct phenotypes in a variety of cancers. It extracts biomarker candidates by employing gene name to link the methylation and gene expression matrices. This tool can assist researchers in examining the statistical significance of predictors/specific genes of interest as well as clustering results.

DEMGD / Dragon Extractor of Methylated Genes in Diseases

Extracts associations between methylated genes and diseases from text abstracts. DEMGD can process data mining into text like PubMed abstracts. The software exploits the text information when it encounters words from these three categories in the same sentence: genes, diseases and methylation. The text-mining methodology is based on a concept of position weight matrices (PWM) and allows the identification of association between concepts.