W1 – Computational Epigenomics

Venue: Eugenides Foundation

Room: Conference Hall

Short Description

Recent advances in single cell sequencing lead to computational challenges in working with this type of uncertain and sparse data. The simultaneous high-throughput measurement of multiple genomewide features in individual cells such as gene expression and epigenetic modifications is in reach, techniques such as NOMe-Seq allow to determine DNA methylation and chromatin accessibility/ nucleosome positioning in single cells. Such new data sets offer the exciting opportunity to directly link mechanisms of epigenetic control to gene expression, without the confounding factor of (unknown) population structures which can affect bulk data sets. While significant advancements were made in processing and evaluating single cell RNA-seq cohorts, much less attention is brought to protocols related to epigenomic data.

While epigenetic effects are widely associated with functional/ phenotypic outcomes, the approaches to integrating epigenetic with other genomic and/or phenotypic data are often ad-hoc and based on heuristics. A systematic characterisation of the underlying mechanisms of epigenetic control of gene expression is still missing. Several studies have highlighted associations between DNA methylation at CpG islands and gene silencing and, more recently, the interaction between histone modifications and gene expression has also attracted considerable attention. Nevertheless, many of these associations are purely correlative, and several other associations can be found, for example between transcription factor binding and histone modifications. Such observations call for more effective computational methods to jointly model multiple epigenomic and other high-throughput data sets, however this poses formidable computational challenges due to the dimensionality, size and complexity of each individual data set. Moreover, mathematical models that describe hypotheses about the mechanisms underlying the formation of DNA methylation patterns and histone modifications will contribute to the understanding of the functioning and interplay between methylation enzymes and chromatin structures. The workshop will review recent progress, and stimulate further work on this topic of central importance in high-throughput bioinformatics.

Target Audience

This workshop targets students and researchers interested in the recent developments on computational epigenetics. It will be a unique opportunity for participants to discover and discuss state of the art models of epigenetic dynamics and methods for working with epigenomic data.


Please send abstract submissions for oral and poster presentations by email to eccb2018@computational-epigenomics.com. Abstracts should include title, author affiliation, contact information and a summary of the work that should not exceed 600 words.


Important dates:

Submission Opening date: June 6th, 2018
Submission Deadline (abstract): July 12th, 2018
Notification: July 19th, 2018
Workshop Date: September 8&9, 2018

Preliminary Program


Saturday, September 8

Session 1 – Modeling of epigenetic dynamics

Chair: Verena Wolf

09:00 10:00 Uwe Ohler (keynote)


10:00 10:30 Gabriele Schweikert

Nonparametric identification of gene-specific spatial histone modifications signatures

10:30 11:00 Coffee/Tea break
11:00 11:30 Alexander Lück

A stochastic model for the formation of spatial methylation patterns

11:30 12:00 Reka Toth

Decipher cell-type composition in cancer epigenetics

12:00 12:30 Laura Cantini

Stabilized independent component analysis outperforms other methods in finding reproducible signals in tumoral transcriptomes

12:30 14:00 Lunch break

Session 2 – Single cell epigenomics

Chair: Yassen Assenov

14:00 15:00 Oliver Stegle (keynote)


15:00 15:30 Coffee/Tea break
15:30 16:00 Carl Hermann

Single-cell transcriptome and chromatin accessibility data reveals inter- and intra-tumor regulatory heterogeneity

16:00 16:30 Ricard Argelaguet

Single-cell multi-omics profiling reveals a dynamic epigenetic landscape of mammalian germ layer specification

16:30 17:00 Andreas Kapourani

Melissa: Bayesian clustering and imputation of single-cell methylomes

17:00 18:30 Poster session
Sunday, September 9

Session 3 – Integrative regulatory genomics

Char: Guido Sanguinetti

09:00 10:00 Marcel Schulz (keynote)


10:00 10:30 Julia Schulze-Hentrich

Plasticity of the epigenome under environmental factors in the pathogenesis of Parkinson’s disease

10:30 11:00 Coffee/Tea break
11:00 11:30 Gwenneg Kerdivel

Deciphering the role of epigenetic modifications in aggressiveness of adrenocortical carcinoma

11:30 12:00 Jon Higham

Analysis of a longitudinal cohort delineates the genetic control of age associated epigenetic changes

12:00 12:30 Juan Castillo Fernandez

Identification of DNA methylation changes associated to ageing outcomes in human population samples

12:30 14:00 Lunch break

Session 4 – Epigenetic regulation of transcription

Chair: Marcel Schulz

14:00 14:30 Florian Schmidt

A supervised method for enhancer identification and linkage to target genes

14:30 15:00 Jing Yang

Time course analysis of open chromatin, promoter interactions and gene expression in primary T-cells implicates causal genes enriched for GWAS SNPs

15:00 15:30 Coffee/Tea break
15:30 16:00 Jan Grau

Prediction of in-vivo transcription factor binding sites in plants

16:00 16:30 Andigoni Malousi

Assembling local genomic signatures of differentially methylated sites from high-throughput analyses

16:30 17:00 Joint discussion: current challenges in computational epigenomics



Yassen Assenov, 

Group Leader, Computational Epigenomics, German Cancer Research Center

Guido Sanguinetti, 

Professor, Computational Bioinformatics, University of Edinburgh

Jordana Bell, 

Senior Lecturer, Epigenomics Research Group, King’s College London

Jörn Walter, 

Professor, Genetics/Epigenetics, Saarland University

Christoph Bock, 

Principal Investigator, Medical Epigenomics Laboratory, 

CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences

Verena Wolf, 

Professor, Modeling and Simulation, Saarland University