Laboratory of Functional and Structural Genomics

Prof. Dariusz Plewczyński

Title	Project Leader	Project period	Project funding
Multiscale spatial reorganization of chromatin in response to replication stress and its role in cellular protection against genomic instability	Dariusz Plewczyński	2021 - 2026	OPUS 19, NCN
The (MSCA)-ITN the Innovative Training Networks program ENHPATHY	Dariusz Plewczyński	2020 - 2023	H2020
Three-dimensional Human Genome structure at the population scale: computational algorithm and experimental validation for lymmphoblastoid cell lines of selected families from 1000 Genomes Project	Dariusz Plewczyński	2017 - 2021	TEAM III, FNP
Genome-wide analysis of DNA double-strand breaks sites in the context of human inter-individual variability and the higher order chromatin organization		2020 - 2021	MINIATURA 4, NCN
Optimisation of HiChIP method to generate high-resolution map of chromatin contacts mediated by cohesin in the human genome	Karolina Jodkowska	2020 - 2021	MINIATURA 4, NCN
Ebola	Dariusz Plewczyński	2019 - 2020	EDCTP
iCell: information processing in living organisms. The role of three-dimensional structure and multi-scale properties in controlling the biological processes in a cell.	Dariusz Plewczyński	2015 - 2019	OPUS, NCN
Virtual High Throughput Screening (vHTS) derivation of a cross-immunity model for the Influenza-A Virus Infections	Dariusz Plewczyński	2014 - 2017	OPUS, NCN

Title	Deadline for applications
Postdoc (Adjunct)(CeNT-58-2023)	15/01/2024
PhD Student CeNT-36.1-2023	01/11/2023
PhD Student (CeNT-36-2023)	15/08/2023
PhD Student (CeNT-5-2023)	28/02/2023
PhD Student (CeNT-56-2022)	15/12/2022
Postdoc (Adjunct)(CeNT-46-2022)	28/09/2022
Adjunct (CeNT-21.2-2021)	30/06/2022
Postdoc (Adjunct) (CeNT-20.1-2021)	08/12/2021
Postdoc (Adjunct) in the Laboratory of Functional and Structural Genomics	31/10/2021
Postdoc (Adjunct) in the Laboratory of Functional and Structural Genomics	31/10/2021
PhD Student in the Laboratory of Functional and Structural Genomics	15/09/2021
PhD Student in the Laboratory of Functional and Structural Genomics	01/04/2021
PhD position in the ENHPATHY Innovative Training Network (Marie Skłodowska-Curie MSCA-ITN Project 860002) – Poland	03/07/2020
PhD student in Laboratory of Functional and Structural Genomics	19/01/2020
Chromatin conformation capture postdoc – experimental work in NGS-based 3D genomics techniques: Hi-C, ChIA-PET and HiChIP	12/10/2019
2 Computational modelling postdocs	12/10/2019
Computational modelling MSc Student	12/10/2019
Technical position	10/10/2019
MSc student in Laboratory of Functional and Structural Genomics	29/04/2019
Adiunkt	13/01/2019
Computational modelling PhD students	01/09/2018
Computational modelling MSc student	01/09/2018
Postdoctoral researchers to work in the Laboratory of Functional and Structural Genomics	01/09/2018

Selected Scientific Discoveries in Laboratory of Functional and Structural Genomics headed by Dariusz Plewczynski, PhD at Centre of New Technologies, University of Warsaw, Poland
In the Laboratory of Functional and Structural Genomics we perform theoretical studies, whose main objective is to analyze and predict the three-dimensional structure of the human genome, and its relation with the genomic diversity of human populations, both natural and pathological. In particular, we investigate structural variants, copy number variants observed in various sub-populations and the groups of patients, and their three- dimensional localization in the structure of the nucleus. We also examine the relationship of the expression
levels of selected genes from their location in three-dimensional space. In addition, we use structural information to enrich the sequential genomic analysis in order to better define the function of selected genomic regions that are important in the context of personalized medicine. For this purpose, first we are developing a variety of large-scale computational tools for analysis of whole genome sequences, the identification of structural variants, determining the statistical significance of the observed number of copies of genomic regions in selected cohorts of patients. Secondly, we evaluate their uniqueness comparing the observed changes with typical and natural genomic diversity that has been cataloged for example in the 1000 Genomes Project Consortium. Thirdly, we infer the biological function of these genomic regions using publicly available databases. Fourthly, we identify unique local three-dimensional environment for selected sites, eg. regulatory ones. In the fifth step, we analyze the impact of structural re-arrangements of those local neighborhoods on the gene expression profiles, which is related to the presence of transcription factories.

Areas For Scientific Synergies:
Three dimensional Genomics, higher order chromatin organization, nucleus, GWAS & SV (deletions, duplications, insertions, inversions, and translocations), Big Data, statistical learning, massive dataset analysis; Computational Genomics and bioinformatics in population-level genomic data for medical applications and fundamental research in Life Sciences, structural and functional analysis of “omics” data; SV identification in human genome using NGS methods; biophysical methods, proteins structure prediction, protein-protein interactions: prediction and analysis and analysis of biomolecule interactions networks; whole-cell modeling;

Laboratory Research Methodology and Achievements
Our research methodology was already tested and validated experimentally for GM12878 cell line only, where the high resolution interaction datasets were available. The underlying biological model with experimental results was published in Cell 2015: http://linkinghub.elsevier.com/retrieve/pii/S0092-8674(15)01504-4 , followed by the detailed description of computational model that was published in Genome Research 2016 http://genome.cshlp.org/content/early/2016/10/27/gr.205062.116.abstract. The web server allowing research community for free access to both the experimental data and computational simulation engine was published in Nucleic Acids Research 2016: http://nar.oxfordjournals.org/content/early/2016/05/16/nar.gkw437.full . In short: ChIA-PET is a high throughput mapping technology that reveals long-range chromatin interactions and provides insights into the basic principles of spatial genome organization and gene regulation mediated by specific protein factors. Recently, we showed that a single ChIA-PET experiment provides information at all genomic scales of interest, from the high resolution locations of binding sites and enriched chromatin interactions mediated by specific protein factors, to the low resolution of non-enriched interactions that reflect topological neighborhoods of higher-order chromosome folding. This multilevel nature of ChIA-PET data offers an opportunity to use multiscale 3D models to study structural-functional relationships at multiple length scales, but doing so requires a structural modeling platform. We development of 3D-GNOME (3-Dimensional GeNOme Modeling Engine), a complete computational pipeline for 3D simulation using ChIA-PET data. 3D- GNOME consists of three integrated components: a graph-distance- based heatmap normalization tool, a 3D modeling platform, and an interactive 3D visualization tool. Using ChIA-PET and Hi-C data derived from human B-lymphocytes, we demonstrate the effectiveness of 3D-GNOME in building 3D genome models at multiple levels, including the entire genome, individual chromosomes, and specific segments at megabase (Mb) and kilobase (kb) resolutions of single average and ensemble structures. Further incorporation of CTCF-motif
orientation and high-resolution looping patterns in 3D simulation provided additional reliability of potential biologically plausible topological structures.

Structural Genomics – to be further developed within proposed TEAM project

1. “An integrated 3-Dimensional Genome Modeling Engine for data-driven simulation of spatial genome organization” Szałaj P, Tang Z, Michalski P, Pietal MJ, Luo OJ, Sadowski M, Li X, Radew K, Ruan Y, Plewczynski D. Genome Res. 2016 Oct 27.
2. “3D-GNOME: an integrated web service for structural modeling of the 3D genome” by Szalaj P, Michalski PJ, Wróblewski P, Tang Z, Kadlof M, Mazzocco G, Ruan Y, Plewczynski D. Nucleic Acids Res. 2016 May 16. pii: gkw437. pmid:27185892

3D-GNOME (3-Dimensional Genome Modeling Engine) is a complete computational pipeline for 3D simulation
using ChIA-PET or Hi-C data. 3D-GNOME consists of four integrated components: i) a graph-distance- based heat map normalization tool, ii) a 3D modeling platform, iii) an interactive 3D visualization tool, and finally iv) a web service which generates 3D structures from 3C data and provides tools to visually inspect and annotate the resulting structures, in addition to a variety of statistical plots and heatmaps which characterize the selected genomic region. Using ChIA-PET and Hi-C data derived from human B-lymphocytes, we demonstrate the effectiveness of the 3D-GNOME in building 3D genome models at multiple levels, including the entire genome, individual chromosomes, and specific segments at megabase (Mb) and kilobase (kb) resolutions of single average and ensemble structures. Further incorporation of the CTCF-motif orientation and high- resolution looping patterns in the 3D simulation provided additional reliability of potential biologically plausible topological structures. The 3D-GNOME web server is freely available at http://3dgnome.cent.uw.edu.pl/.

3. “CTCF-Mediated Human 3D Genome Architecture Reveals Chromatin Topology for Transcription” Tang Z, Luo OJ, Li X, Zheng M, Zhu JJ, Szalaj P, …, Wilczynski GM, Plewczynski D, Li G, Ruan Y. Cell 2015, Dec 17;163(7):1611-27. Epub 2015 Dec 10.

Spatial genome organization and its effect on transcription remains a fundamental question. We applied an advanced chromatin interaction analysis by paired-end tag sequencing (ChIA-PET) strategy to comprehensively map the higher-order chromosome folding and specific chromatin interactions mediated by the CCCTC-binding factor (CTCF) and RNA polymerase II (RNAPII) with haplotype specificity and nucleotide resolution in different human cell lineages. We find that the CTCF/cohesin-mediated interaction anchors serve as structural foci for spatial organization of constitutive genes concordant with the CTCF-motif orientation, whereas RNAPII interacts within these structures by selectively drawing cell-type- specific genes toward CTCF foci for coordinated transcription. Furthermore, we show that the haplotype variants and allelic interactions have differential effects on chromosome configuration, influencing gene expression, and may provide mechanistic insights into functions associated with disease susceptibility. The 3D genome simulation suggests a model of chromatin folding around chromosomal axes, where CTCFs are involved in defining the interface between condensed and open  compartments for structural regulation. Our 3D genome strategy thus provides unique insights into the topological mechanism of human variations and diseases.

4. “A global reference for human genetic variation” by 1000 Genomes Project Consortium, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA, Abecasis GR. Nature. 2015 Oct 1;526(7571):68-74.

The 1000 Genomes Project set out to provide a comprehensive description of common human genetic variation by applying whole-genome sequencing to a diverse set of individuals from multiple populations. This publication reports the completion of the project, having reconstructed the genomes of 2,504 individuals from 26 populations using a combination of low-coverage whole-genome sequencing, deep exome sequencing, and dense microarray genotyping. They characterized a broad spectrum of genetic variation, in total over 88 million
variants (84.7 million single nucleotide polymorphisms (SNPs), 3.6 million short insertions/deletions (indels), and 60,000 structural variants), all phased onto high-quality haplotypes. This resource includes ~99% of the SNP variants with a frequency of >1% for a variety of ancestries. The distribution of genetic variation across the global sample is provided, and discusses the implications for common disease studies. Our (D. Plewczynski) contribution was structural variants (SV) identification using combined aCGH arrays and NGS sequencing tightly coupled with computational meta-approach. The set of tools (including meta-caller for SV identification using combined several independent methods and machine learning was developed at Charles Lee laboratory at JAX GM. We participated in the early phase of this project, providing initial results and working independently further within this area by focusing on application of deep learning algorithms for SV calling and GWAS studies.

Functional Genomics – tools and algorithms developed within iCell (2015-2018) and Influenza (2013-2016) National Research Centre OPUS grants

5. “Detecting reliable non interacting proteins (NIPs) significantly enhancing the computational prediction of protein–protein interactions using machine learning methods" by Srivastava,A, Mazzocco,G, Kel A, Wyrwicz LS and D. Plewczynski Mol BioSyst. 12(3):778-85 (2016).

6. “Ensemble learning prediction of protein-protein interactions using proteins functional annotations” by Saha, Zubek, Klingström, Forsberg, Wikander, Kierczak, Maulik and D. Plewczynski Mol BioSyst. 10(4):820-30. (2014).

Protein-protein interactions (PPI) are important for the majority of biological processes. A significant number of computational methods have been developed to predict protein-protein interactions using protein sequence, structural and genomic data. Vast experimental data is publicly available on the Internet, but it is scattered across numerous databases. This fact motivated us to create and evaluate new high-throughput datasets of interacting proteins, such as DIP, MINT, BioGRID and IntAct. Then we constructed descriptive features for
machine learning purposes based on data from Gene Ontology and DOMINE. Thereafter, four well-established machine learning methods: Support Vector Machine, Random Forest, Decision Tree and Naïve Bayes, were used on these datasets to build an Ensemble Learning method based on majority voting. In cross-validation experiment, the sensitivity exceeded 80% and the classification/prediction accuracy reached 90% for the Ensemble Learning method. We extended the experiment to a larger and more realistic dataset maintaining
sensitivity over 70%. Finally, we further improved those results by constructing a unique validation approach aimed to at collecting reliable non-interacting proteins (NIPs). The curated negatives and positives subsets were used for PPI classification, leveraging the prediction capabilities of well-established machine learning methods. Our best classification procedure displayed specificity and sensitivity values of 96% and 98%, respectively, surpassing the prediction capabilities of other methods, including those trained on gold standard datasets. We showed that the PPI/NIP predictive performances could be considerably improved by focusing on data preparation. The predicted PPI networks were analyzed in terms of their topological and geometrical properties.

7. “Consensus classification of Human Leukocyte Antigens class II proteins” by I. Saha, G. Mazzocco and D. Plewczynski. Immunogenetics 65(2):97-105 (2013);

Class II human leukocyte antigens (HLA II) are proteins involved in the human immunological adaptive response by binding and exposing some pre-processed, non-self peptides in the extracellular domain in order to make them recognizable by the CD4+ T lymphocytes. However, the understanding of HLA-peptide binding interaction is a crucial step for designing a peptide-based vaccine because the high rate of polymorphisms in HLA class II molecules creates a big challenge, even though the HLA II proteins can be grouped into supertypes, where members of different class bind a similar pool of peptides. We performed the supertype classification of 27 HLA II proteins using their binding affinities and structural-based linear motifs to create a stable group of supertypes. For this purpose, a well-known clustering method was used, and then, a consensus was built to find the stable groups and to show the functional and structural correlation of HLA II
proteins. Thus, the overlap of the binding events was measured, confirming a large promiscuity within the HLA II-peptide interactions. Moreover, a very low rate of locus-specific binding events was observed for the HLA-DP genetic locus, suggesting a different binding selectivity of these proteins with respect to HLA-DR and HLA-DQ proteins. Secondly, a predictor based on a support vector machine (SVM) classifier was designed to recognize HLA II-binding peptides.

Public Health Relevance of proposed research project 
It is becoming increasingly recognized that higher-order genome organization crucially influences gene regulation, cell function and ultimately human health. However, the ability to investigate these relationships will depend on technological advances that enable a more precise, integrated view of genome structure and function. This proposal seeks to address this challenge through the development of a 3Depigenomic platform that will provide a powerful tool for studying genomic structure in space (3D) and time (4D).