Quick Links

Seminars for the Systems Biology Training Program

November 2014

Speaker: Michael Hoffman, Princess Margaret Cancer Centre
When: Friday Nov. 21st, 11:00 - 12:00pm
Where: Wong Building, room 1050

Title: Semi-automated genome annotation and an expanded epigenetic alphabet

Abstract: First, we will discuss Segway, an integrative method to identify patterns from multiple functional genomics experiments, discovering joint patterns across different assay types. We apply Segway to ENCODE ChIP-seq and DNase-seq data and identify patterns associated with transcription start sites, gene ends, enhancers, CTCF elements, and repressed regions. Segway yields a model which elucidates the relationship between assay observations and functional elements in the genome. Second, we will discuss a new method to discover transcription factor motifs and identify transcription factor binding sites in DNA with covalent modifications such as methylation. Just as transcription factors distinguish one standard nucleobase from another, they also distinguish unmodified and modified bases. To represent the modified bases in a sequence, we replace cytosine (C) with symbols for 5-methylcytosine (5mC), 5-hydroxylmethylcytosine (5hmC),
5-formylcytosine (5fC), and 5-carboxylcytosine (5caC). Similarly, we adapted the well-established position weight matrix model of transcription factor binding affinity to an expanded alphabet. We created an expanded-alphabet genome sequence using genome-wide maps of 5mC, 5hmC, 5fC, and 5caC in mouse embryonic stem cells. Using this sequence and expanded-alphabet position weight matrixes, we identified cis-regulatory modules that we believe are active only in the presence of cytosine modifications.

All are welcome!



2013 Canadian Cancer Research Conference

3-6 November 2013, Toronto, ON

International Keynote Speakers Include

Johann de Bono, Institute of Cancer Research, UK

Carlos Caldas, Cancer Research UK Cambridge Institute, UK

Stephen Friend, Sage Bionetworks, USA

Judy Garber, Dana-Farber Cancer Institute, USA

Peter Jones, USC Norris Comprehensive Cancer Center, USA

And a host of your colleagues from across Canada



for further information




November 2013

November 8, 2013 at 1:30pm

Karp Amphitheatre, room 501, Goodman Cancer Research Centre

Dr. William Dalton, PhD, MD

H Lee Moffitt Cancer Centre, Tampa, Florida, USA



October 2013

October 16, 2013 at 12pm

McIntyre Medical Building, Martin Theatre, room 504

Therese Sorlie, PhD

Group Leader, Department of Genetics, Institute for Cancer Research, Oslo University, Norway




September 2013

September 30, 2013 at 15:00

Room: Strathcona M1

Philippe Sanseau, PhD

Head of Computational Biology at GlaxoSmithKline, UK



April 2012

April 18, 2012

Sylvain Ethier

 PhD Systems Biology student, Dept. of Biochemistry, McGill University McGill University

 Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

TITLE: Lesson from Estrogen Receptor-regulated genes


Orchestrated regulation of gene expression is done by integrating physiological and environmental cues. This regulation is lead by numerous regulatory events that control both the repression and activation of genes. Estrogen nuclear receptors (ER) are a class of transcription factor that were shown to be overexpressed in breast cancer. When bound by its ligand, estradiol, ER coordinates a gene expression program and modification of chromatin structure which implicates long-range interaction by FoxA1.  This thesis addresses long-range promoter-enhancer relationship in a highly dynamics context using a high-resolution chromatin structure technique refered to as Chromosome Conformation Capture Carbon Copy (5C).   This work gives insight on some questions regarding gene regulation, such as

1)  Can chromatin structure be responsible for cyclical behavior of gene expression ?  2) Can coordination of ER-regulated gene be modulated by transcription factories ?  3) Can distance between a promoter and his enhancer be a regulation mechanism for gene expression ?


April 11, 2012

Daniel Cooper

 PhD Systems Biology student, Dept. of Physics, McGill University McGill University

 Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

TITLE: Toward Imaging Neuronal Membrane Potentials

Abstract: Robust imaging of neuronal membrane potentials in real time would facilitate valuable insights into the fundamental dynamics of neural excitation and the pathophysiology of many debilitating diseases. Many efforts toward developing voltage-sensitive probes have focused on the modification of membrane channels that are themselves complicit in many of the conditions under study. We have explored several avenues by which to create such an endogenous membrane probe, though have been confounded by complications that, while ultimately frustrating our efforts, may be of some value to posterity.


April 4, 2012

Evelyn Zastepa

 PhD Student, Department of Neuroscience, McGill University

 Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

Aberrant transcriptional program regulating naive CD4 T-cell activation defines a progressive multiple sclerosis subgroup

Introduction: Multiple sclerosis (MS) is a chronic inflammatory neurodegenerative and demyelinating disease of the central nervous system. Peripheral immune mechanisms drive disease in relapsing remitting MS (RRMS), whereas contribution of immune mechanisms to progressive MS, including secondary progressive MS (SPMS) and primary progressive MS (PPMS), is uncertain. Naïve CD4 T-cells initiate autoimmune responses. Methods: To clarify the role of these cells in progressive MS, we studied their gene expression profiles/osb in 24 MS patients (19 SPMS, 5 PPMS) and 14 healthy controls (HC) using microarrays. Key genes and pathways were validated by surface protein expression and functional studies.  Results: Hierarchical clustering, based on most variable genes, separated progressive MS patients into two subgroups, which we termed P-1 and P-2. Most differentially expressed genes between P-1 and HC were upregulated in P-1 and enriched for immune response; whereas in P-2, most were downregulated compared to P-1; suggesting immune mechanisms are active only in P-1 patients. Further pathway analysis was focused on SPMS patients in subgroup P-1. Analysis suggests altered T-cell receptor (TCR) and Toll-like receptor (TLR) signalling dysregulates the naive CD4 T-cell NF-κB cascade in P-1 SPMS patients. We report a 5 gene T-cell activation transcriptional signature specific for P-1 SPMS. Interestingly, P-1 SPMS patients had a shorter RRMS duration than P-2 patients, suggesting a more active/rapid disease course. Protein expression studies support the microarray findings. Further, P-1 SPMS patients showed a differential response to TCR stimulation. Conclusions: Our transcriptional signature defines for the first time a subgroup of SPMS patients having rapid RRMS progression and T-cell immune dysregulation, who may benefit from immunomodulatory treatments.


March 2012

March 28, 2012

Deena Gendoo

PhD 5 Bioinformatics Candidate, SYPA Chair, PrioNet, Biology, McGill University

Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

Bioinformatic Sequence and Structural Analysis for Amyloidogenicity in Prions and Other Proteins

Detection of amyloidogenic peptides or domains in proteins is of paramount importance towards understanding their role in amyloidosis in conformational diseases. This presented work explores different methods towards detection and prediction of amyloidogenic peptides using a variety of bioinformatic analytical methods. Bioinformatic analysis of secondary structural changes is employed to determine whether classes of structurally ambivalent peptides, mainly discordant and chameleon sequences, are efficient predictors of amyloidogenic segments. This analysis elucidates statistical relationships between discordance, chameleonism, and amyloidogenicity across a database of protein domains (SCOP), a subset of amyloid-forming proteins, and the prion family. The presented results stress upon the limitations of these peptides as predictors of amyloidogenicity, and raise issues on the predictive power that can be reaped from secondary structure prediction methods.  In another bioinformatic approach, detection of conformationally variable segments in tertiary structures of PrP globular domains has been performed using Principal Component Analysis. This technique succeeded in identifying five conformationally variable domains within PrP, and ranking these subdomains by their ability to differentiate PrPs based on non-local structural response to pathogenic mutation and prion disease susceptibility. The presented results are corroborated by previous observations from experimental methods and molecular dynamic simulations, suggesting that this approach serves as a fast and reliable method for detection of potential amyloidogenic segments in amyloid-forming proteins. Finally, a structural, functional, and evolutionary bioinformatic analysis is conducted to assess the prevalence of the first experimentally verified amyloid fibril fold in nature, and whether this fold can serve as a prototype for other amyloid-forming proteins. The results indicate a limited scope of this fold in amyloid-forming proteins and across the protein universe, and have implications on future identification of amyloid-forming proteins that share this fold. The research presented lays the groundwork for similar and future studies that can be conducted on known amyloid-forming proteins or potential amyloidogenic candidates.

March 21, 2012

Dr. Monique Zetka

Associate Professor, Department of Biology, McGill University

Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

Regulation of meiotic chromosome pairing in C. elegans

During meiosis, duplicated chromosomes must find and align along their lengths with their homologous partners. These complex events require forces to move chromosomes together for homology assessment and to separate nonhomologous chromosomes from one another to continue the homology search. SUN and KASH-domain protein pairs play a conserved role by establishing transient linkages between chromosome ends and cytoskeletal forces across the intact nuclear envelope. In C. elegans, we show that the establishment of these linkages and chromosome pairing requires the polo kinase PLK-2. Once paired, most organisms polymerize a proteinaceous structure known as the synaptonemal complex (SC) between the chromosomes. However, SC forms readily between nonhomologous sequences and an outstanding question is how SC polymerization is controlled to occur only between homologs. We show that the highly conserved type II chaperonin CCT has a nuclear role in mediating this process by assisting in the assembly of meiotic chromosomes competent for SC formation once paired. These results provide key insights into the regulation of homolog pairing.


March 7, 2012

 Dr. Guillaume Bourque

Assoc. Professor, Department of Human Genetics, McGill University

Director of Bioinformatics, McGill University & Genome Quebec Innovation Center

Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

The role of genomic repeats in gene regulation

 Next-generation sequencing (NGS) technologies (e.g. ChIP-Seq, RNA-Seq) are now supplanting array-based technologies because of their accuracy, comprehensiveness and cost. Notably, these technologies enable an unbiased look at the functional contributions of the genome including the contributions of repetitive regions. We will present results that demonstrate the ubiquitous role that play repeats in gene regulation. In particular, we will show that species-specific transposable elements have been an important source of new regulatory elements and have contributed more than 20% of the binding sites of key transcription factors in human ES cells. We will also present results from the analysis of DNaseI hypersensitivity and ChIP-Seq ENCODE datasets in normal, embryonic and cancer cell lines that demonstrate that in all cell types between 8% and 12% of the open chromatin regions have been contributed by transposable elements with a significant over-representation of endogenous retroviral (ERV) repeat families. Interestingly, we further show that hundreds of these transposons-derived sequences are activated in a cell-type specific manner. We are also able to confirm that genes near repeat instances containing hypersensitive sites in a particular cell type tend to be more differentially expressed. Although we find that most of this tissue-specific activity on transposon-derived sequences is detected in embryonic stem cells and cancer cell lines, a few repeat families are also open in distinct differentiated cell types. For instance, close to 30% of the instances of the LTR2B ERV subfamily are only open in lymphocyte-related and in renal epithelial cells. Given that various repeat families are highly polymorphic across individuals, characterizing their regulatory activity is likely to be critical to understand functional variation and susceptibility to disease.


February 2012

February 15, 2012

Dr. Alvin Shrier

Director of the Cell Information Theme and the Life Sciences Imaging Facility Physiology, McGill University

Time: 2:30 pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

Title: hERG potassium channel surface expression and cardiac rhythms

The potassium current IKr encoded by the hERG gene is important for controlling cardiac excitability and action potential repolarization. Mutations in hERG give rise to the congenial long QT syndrome (LQTS) that is associated with cardiac arrhythmias and sudden death. Moreover, drug binding to the hERG channel represents a serious clinical problem and consequently the FDA requires that all new drugs must be screened against hERG. We have been investigating the nature of the maturation of the hERG channel protein and it’s trafficking to the cell membrane. In addition, we have been studying the nature of the quality control mechanism that determines the stability of the hERG channel once it has arrived at the cell membrane. Our studies reveal a number of new chaperones involved in maturation and evidence for a peripheral quality control system. In studies of the functional effect of having a reduced hERG expression, we have conducted studies of the dynamics of excitability of cultured cardiac cells.  We find that hERG block induces irregular and even chaotic rhythms that may underlie clinical cardiac arrhythmias.


February 8, 2012

Mathieu Rousseau

PhD student, Computer Science, McGill University

TIME:   2:30pm

PLACE:  Room 1034, McIntyre Medical Building, 3655 promenade Sir William Osler, Montreal

Chromatin organization plays an important role in regulating gene expression through the action of long-range enhancers and insulators. The effect of these proteins are made possible by the dynamic higher-order folding and re-organization of chromatin within the nucleus of a cell. One recent technology for determining the structure of DNA is Chromosome Conformation Capture Carbon Copy (5C) that produces a pairwise distance matrix between specific points within the genome. These data are noisy and require robust computational modeling and analysis techniques for interpretation.

In this thesis, we look at modeling 5C data and develop computational approaches to determine chromatin three-dimensional structure and look to provide insight into the role of chromatin structure with respect to gene expression and function. This work is divided into three results, 1) a novel model for 5C data and a Markov chain Monte Carlo simulation-based approach for determining structure, 2) a timecourse experiment designed to address the open/closed-on/off controversy regarding the role of chromatin structure in gene expression, and 3) a collaborative project to determine if chromatin structure can differentiate between types of MLL-fusion leukemias.


April 2011

April 27, 2011
2:30 pm

"Systems epidemiology - some experiences with the Norwegian Women and Cancer postgenome study”

Eiliv Lund

Professor, Institute of Community Medicine, University of Tromsø, Norway

Trottier Building, room 1080, 3630 University St.

Systems epidemiology can be considered as an attempt to implement functional analyses into the common prospective design. Functional analyses cover transcriptomics (mRNA, miRNA) and epigenetics (methylation) from both blood and different tissues. Some of these techniques can be run in available material, some need new biological sampling. The expansion of the information available through these methods has created a challenge for the analyses both in terms of laboratory analyses, statistical analyses and interpretations. At the same time its success will mirror the current dichotomy in cancer research between epidemiology and basic genetic research. With this as a background the design and current analytical approaches of the Norwegian Women and Cancer postgenome cohort will be discussed.

March 2011

March 30, 2011
2:30 pm

"Mitochondrial DNA mutations in human skeletal muscle: cellular and physiological effects”

Tanja Taivassalo
Assistant Professor, Department of Kinesiology, McGill University

Trottier Building, room 1080, 3630 University St.

Mitochondria are essential to human life as they are the body’s primary source of energy. They form the final link of the oxygen cascade; a coordinated pathway describing the transport of oxygen through the lungs, heart, blood vessels to inside the cell for utilization by mitochondria in the process of energy production. These unique organelles contain multiple copies of their own DNA which are critical in their formation and function; a mutation in mitochondrial DNA can result in their dysfunction. This is dramatically demonstrated in patients born with primary mitochondrial disease (PMD), a progressive condition where accumulation of the mutation in muscle decreases cellular energy production, resulting in debilitating fatigue and quality of life. Since discovery of the first mutation in mitochondrial DNA just 2 decades ago, over 200 mutations have been described to cause PMD. Less progress has been made in linking the level of mutation and degree of mitochondrial impairment to physiological and functional disability. Interestingly, mutations in mitochondrial DNA have been detected in muscle of healthy, elderly individuals, fostering speculation that damage-induced mutations accumulate over time, resulting in loss of muscle mass, function and physical capacity. The molecular similarities and declining physiological function make PMD an ideal model to investigate the mitochondrial theory of human aging. An understanding of mechanisms underlying mutant mitochondrial DNA accumulation in muscle and their impact on cellular energy production and physiological functioning is essential if we are to detect, treat and ultimately prevent the effects of mitochondrial impairment in conditions of human disease and aging. This presentation will describe the impact of mutations in skeletal muscle mitochondrial DNA on cellular and physiological function, highlighting the pivotal role they play in regulating the oxygen transport cascade. Furthermore, exercise-based strategies aimed at increasing levels of normal mitochondria in skeletal muscle will be described. As will be discussed, an integrative approach to the scientific study of mitochondrial DNA mutations and their impact is necessary, and includes molecular, cellular and whole-body physiological assessment.


MonBUG Seminar
March 10, 2011
4:00 pm

"BLASTing faster than light: compressive genomics in the next-gen world "

Michael Baym
Systems Biology Department, Harvard Medical School

Leacock Building, Room 232, McGill University, 855 Sherbrooke St.W,Montreal

The last several years have seen an unprecedented increase in DNA sequencing capability. The resulting accumulation of data has grown so fast as to far outstrip advances in computational or even storage capabilities. Fundamental algorithmic improvements are necessary if we are to truly make use of these new technologies. In this talk, we present an extension of BLAST on large datasets that not only is faster, but scales sublinearly in the size of the genomic library. More broadly, we introduce compressive genomics, an approach to these increasingly large data sets that allows analysis to be performed directly on the compressed data.

February 2011

MonBUG Seminar
Friday, February 11, 2011
4 pm

"Probabilistic Graphical Model for Protein Structure Prediction"

Jinbo Xu
Toyota Technological Institute, U. of Chicago

Leacock Building, Room 232, McGill University, 855 Sherbrooke St. West, Montreal

If we know the primary sequence of a protein, can we predict its three-dimensional structure by computational methods? This is one of the most important and difficult problems in computational molecular biology and has tremendous implications for protein functional study and drug discovery. Existing computational methods for protein structure prediction can be broadly classified into two categories: template-based modeling (i.e., protein threading/homology modeling) and template-free modeling (i.e., ab initio folding). Template-based modeling predicts structure of a protein using experimental structures in the Protein Data Bank (PDB) as templates while template-free modeling predicts protein structure without depending on a template. This talk will present new probabilistic graphical models for knowledge-based protein structure prediction. In particular, this talk will present a regression-tree-based Conditional Random Fields (CRF) method for template-based modeling and a Conditional Random Fields/Conditional Neural Fields (CRF/CNF) method for template-free modeling. Experimental results indicate that our template-based method performs extremely well, especially on hard template-based modeling targets and our template-free method is also very promising for mainly-alpha proteins.


January 19th, 2011
2:30 pm

"Looking at interactions from several angles"

Celia Greenwood
Associate Professor, Lady Davis Institute for Medical Research, Jewish General Hospital McGill University, Department of Oncology, and Department of Epidemiology, Biostatistics and Occupational Health

Trottier Building, Room 1080, 3680 University, Montreal 1

Most human genetic studies that aim to identify genetic risk factors for disease have low statistical power to detect interactions. In this talk, I’ll discuss interactions in general, and then present two different statistical approaches for trying to understand how different factors combine together to influence disease risk. The first approach is based on a Bayesian hierarchical model, and the second is a kernel-based data integration approach. The talk will describe these methods in a general, conceptual way.


October 27, 2010
2:30 pm

"The importance of promoter regions in the evolution of novel regulatory pathways"

Troy Ruths
PhD graduate student, Computer Science Department, Rice University, Houston, Texas

Trottier Building, Room 1100, 3630 University, Montreal

A regulatory pathway is defined by a set of genes that interact temporally using transcription factors and transcription factor binding sites, or short nucleotide signatures usually only around 10 bp in length. These pathways define the behavior of cellular mechanics, embryonic development, and cell differentiation. The region of a genome that can harbor transcription factor binding sites for a given gene scales with effective population size, spanning from less than a kilobase in prokaryotes (large populations) to more than a megabase in mammals (small populations); however, a satisfactory explanation for such a scaling between promoters and population size remains unknown. In this talk I will present my work on the effect of population size and promoter region length on the time to evolve a novel regulatory pathway. By developing a computational population-genetic model of regulatory pathway evolution (which we will cover in this talk), I investigated how the length of regulatory promoter regions can either increase or decrease the time to evolve a novel, optimal pathway depending on the population size. Results show that the minimal time for large genomes occurs in small populations and for small genomes it occurs in large populations, a pattern that closely resembles the natural order and one that elucidates the paradox of uncorrelated organism complexity and gene count. Since regulatory pathways are crucial for organism survivability, this pattern gleans insight into the success of the eukaryotic genome and the usefulness of expanding regions of non-coding DNA in complex organisms. Furthermore, these results stand regardless of the regulatory pathway function, and so they provide new insight and theory on the role of neutral evolution in shaping these enigmatic, yet crucial, regions of the genome.


October 13, 2010
2:30 pm

"A Novel Approach for Comparative Genomics & Annotation Transfer"

Eric Rivals
Researcher in computer science and bioinformatics (Directeur de recherches du CNRS)

Trottier Building, Room 1100, 3680 University, Montreal

With the rapid development of sequencing techniques, the situation where a newly sequenced genome needs to be annotated using available (eventually incomplete or not fully assembled) genomes from close species should become more prevalent in the future. We propose a novel approach to genome comparison and use it to develop a system that transfers annotations between the compared genomes. Besides features' sequence similarity, it accounts for the synteny it detects across multiple genomes. The underlying idea is to partition a focus genome according to its pairwise similarities with the other compared genomes. The question is formulated as searching for the intervals that are shared across all genomes under consideration, and maximal in length. If a genomic region is covered by at least one interval, it is conserved across all genomes, and the number of such intervals tells how many multiple local alignment are potentially available with the other genomes. Our algorithm partitions the genome into regions following two criteria: 1/ being shared or unshared across all genomes, 2/ offering a unique or several alignment possibilities. The annotation transfer procedure crosses the focus genome's annotations with these regions and automatically derives the possible alignments for each feature. All features falling entirely in a region offering only one alignment possibility are declared as potentially transferable, and the user may interactively select among those according to various criteria: alignment's percent of identity, feature class, etc. Our approach allows to compare and pre-annotate unfinished genomes, as well as assemblies of Next Generation Sequencing data. We implemented these procedures in an efficient and flexible tool, named QOD, equipped with a user-friendly graphical interface


September 29, 2010
2:30 pm

"Interactive Visualizations of Network Data"

Michael McGuffin
Assistant Professor, École de technologie supérieure, Montreal

Trottier Building, Room 1100, 3680 University, Montreal 1

This talk will be in two parts. The first presents a bimanual popup widget called the hotbox that can be used to invoke up to thousands of commands with fast, gestural-like input to manipulate a network visualization, while using only 2 buttons for input. We demonstrate how this interface can be applied to the visualization of a protein-protein interaction network. For more information
The second part, to be published later this year, presents a multidimensional visualization approach to network visualization, using a hybrid between parallel coordinates and a scatterplot matrix, as well as a novel popup widget called the FlowVizMenu. This research is joint work with Christophe Viau, Igor Jurisica, and Yves Chiricota. Future directions involving multitouch user interfaces and other possibilities will also be discussed. A 19-inch, 10-finger multitouch display will also be available for a demonstration after the talk.


September 1,2010

"Computational Models for RNA structures"

Synopsis: Ribonucleic acids (RNAs) have emerged as one of the most important biomolecules, playing key roles in various aspects of the gene transcription and regulation processes. To achieve their functions, RNAs use sophisticated structures which are determined by their sequences. The development of sound theoretical models is required to analyze and predict these structures. The talks of this seminar series aim to provide a fresh and comprehensive overview of the most recent and innovative frameworks developed in this field. Audience: This symposium is intended for mathematics, physics and computer science researchers, grad students and senior undergrads interested in the developement and application of mathematical and computational methods to structural biology. Biologists and biochemists interested in the most recent theoretical frameworks developed for RNA structures are also encouraged to attend. The biological background necessary for the understanding of the talks will be provided during the seminars.

Time: 2:30- 6:00 pm
Place: Room 1100 and 3120, Trottier Building, 3680 University st.
Contact: Jérôme Waldispühl

May 2010

May 25, 2010
10:00 AM

"Gene expression profiling of late side effects in breast cancer survivors treated with radiotherapy"

Hege Landmark-Hoyvik,Department of Genetics,Institute for Cancer Research Oslo University Hospital Radiumhospitalet,Norway

Room 501, Goodman Cancer Centre,McGill Life Sciences Complex, 1160 Pine ave. West, Montreal, H3A 1A3

Abstract: Advances in detection and treatment of breast cancer have lead to an increasing number of cancer survivors. In consequence, late effects and optimal quality of life have become new important end points in cancer care. Fatigue and radiation-induced fibrosis (RIF) are some of the most frequent complaints among breast cancer survivors. To explore the biological processes underlying both chronic fatigue and fibrosis in a patient cohort, genome-wide expression analyses were performed on whole blood samples from breast cancer survivors. The results confirm that inflammatory process is involved in chronic fatigue pinpointing more precisely the involvement of plasma- and lymphocyte B cells. Substantial differences in blood gene expression were also observed between the fibrotic and non-fibrotic survivors, partly implicating a deregulation of the TGFβ1 pathway during the maintenance phase of fibrosis.

April 2010

April 23, 2010
2:30 pm

"Functional Genomics in Stem Cells, Evolution and Cancer"

Dr. Guillaume Bourque, PhD, Genome Institute of Singapore

Room 26, Leacock Building

Abstract: In the first part of this talk, I will describe a study where we generated occupancy maps for three key regulatory proteins (OCT4, NANOG and CTCF) together with knockdown expression experiments (OCT4) in human and mouse Embryonic Stem cells. Using this data we found that the binding profiles/osb of OCT4 and NANOG were drastically different between the two species with only ~5% of the regions homologously occupied. Moreover, I will show that speciesspecific transposable elements have contributed up to 25% of the regulatory sites in both lineages and have wired new genes into the core regulatory network of human ES cells. In the second part of this talk, I will look at the forces shaping vertebrate genome architectures and present a new algorithm called EMRAE that can predict a wide-range of rearrangement events in the ancestry of a group of species. Applying this tool to a dataset with six genomes (human, chimpanzee, rhesus macaque, mouse, rat, and dog), I will show that regions of high sequence identity have been associated with rearrangement events throughout the mammalian phylogeny. Finally, in the third part of this talk, I will show how new paired-end-tag sequencing strategies can efficiently provide the map of all structural changes in a given genome. Applying this technique to normal samples, cancer tumours and cancer cell lines and performing a comparative analysis can reveal different patterns of mutations and help identify important aberrations.

March 2010

March 22, 2010
10:00 AM

"Using focused computational models to study endoplasmic reticulum pathways in human disease"

Sara Gosline, McGill University

Room 501, Goodman Cancer Centre,McGill Life Sciences Complex, 1160 Pine ave. West, Montreal, H3A 1A3

Background: The endoplasmic reticulum (ER) is a cellular organelle responsible for lipid biosynthesis, protein folding, and drug detoxification. One third of all cellular proteins are folded and assembled in the ER, including most membrane-bound and secreted proteins that are responsible for inter-cellular signaling. The importance of ER-resident processes is reflected in its link to many diseases such as Cystic Fibrosis,neurodegenerative diseases, diabetes and many types of cancer. However, the unique and dynamic environment of the ER has made it difficult to study using typical cell-wide computational or biochemical approaches. In this talk I will present our systematic approach to studying the ER. Through the development of ER-specific computational and biochemical approaches we have been able to better study the role of the organelle in human disease. More specifically, I will discuss the use of computational models of protein domain conservation to characterize the evolution of the Unfolded Protein Response (UPR) that has led to exciting findings in Leishmania, an organism that causes the deadly Leishmaniasis infection. I will also discuss the use of ER-specific gene signatures to uncover the role of the organelle in breast cancer subtypes and the use of ER-specific methods to better understand ER biology.

December 2009

MonBUG Seminar
December 10, 2009
6 to 9 pm

"The role of evolutionarily conserved ligand-interacting binding-site residues"

Rafael Najmanovich
Assistant Professor, Biochemistry Dept, Université de Sherbrooke

Leacock Building, Room 232, McGill University, 855 Sherbrooke St.W,Montreal

Evolutionarily conserved ligand-interacting binding-site residues are thought to be important to satisfy physico-chemical binding constraints. Recently we demonstrated that non-homologous proteins that evolved to bind similar ligands contain highly dissimilar patterns of conserved ligand-interacting binding-site residues. While the importance of conserved residues is unquestionable, these results suggest that conserved residues may play extra roles. We suggest that conserved residues may play the additional role of preventing the promiscuous binding of similar molecules present in the cellular milieu. Different patterns of conservation would reflect distinct cellular contexts. In support of this hypothesis, we created a dataset (and associated web-interface) of proteins with known structures containing binding site mutations in which both the wild type and mutant were crystallized with the ligand (as well as sometimes in the Apo form) demonstrating that the mutation does not prevent ligand binding. The dataset includes over 5000 entries containing between one and four mutations on residues with varying degrees of conservation in contact with ligands with varying levels of cognate similarity. Due to the experimental bias towards mutations on highly conserved residues, there are numerous cases of often-drastic changes on highly conserved residues bound to cognate ligands. If such drastic mutations do not prevent binding, the conserved residues in question must have a different essential function, including that of preventing promiscuity among other possibilities.

December 2009

December 9th, 2009
2:30 - 3:30 pm

"Functional genomics in cancer and in stem cells"

Guillaume Bourque
Genome Institute of Singapore

Room 903, McIntyre Building, 3655 Promenade Sir-William-Osler, Montreal, Quebec

Next-Generation Sequencing has enabled a wealth of new applications. In this talk, I will show how new paired-end-tag sequencing strategies can efficiently provide the map of all structural changes in a given genome. Applying this technique to normal samples, cancer tumors and cancer cell lines and performing a comparative analysis can reveal different patterns of mutations and help identify important aberrations. Next, I will also look at some of the sequenced-based approaches that are used to explore transcriptional regulatory networks. Specifically, I will describe a study where we generated occupancy maps for three key regulatory proteins (OCT4, NANOG and CTCF) together with knockdown expression experiments (OCT4) in human and mouse Embryonic Stem cells. In this study we found that the binding profiles/osb of OCT4 and NANOG were drastically different with only ~5% of the regions homologously occupied. We also showed that species-specific transposable elements have contributed up to 25% of the regulatory sites in both lineages and have wired new genes into the core regulatory network of human ES cells.


November 26th, 2009
1:00 - 2:30 pm

"Extracting information from biological networks"

Leonid Chindelevitch
Massachusetts Institute of Technology

Room 2120, Trottier Building, 3630 University Street, Montreal, Quebec H3A 2B2 1

In this talk, I will describe different types of biological networks and the information they can provide. I will first present a novel algorithm, PISwap, for computing global pairwise alignments of protein interaction networks, based on a local optimization heuristic that has previously demonstrated its effectiveness for a variety of other NP-hard problems, such as the Traveling Salesman Problem. Some interesting insights into the evolutionary history of the compared species will also be discussed. If time allows, I will briefly mention ongoing work on metabolic and signaling networks.

November 2009

MonBUG Seminar
November 12, 2009
6 to 9 pm

"How perfect can protein interactomes be?"

Christian Landry
Faculté de Médecine Université de Montréal

Leacock Building, Room 232, McGill University, 855 Sherbrooke St. West, Montreal

Evolutionary theory tells us that biological systems need not be optimized and may very well accumulate nonfunctional elements. Mutational and demographic processes contribute to the cluttering of eukaryotic genomes and transcriptional networks with “junk” DNA and spurious DNA binding sites. Here, I question whether such a notion should be applied to protein interactomes- that is, whether these protein interactomes are expected to contain a fraction of nonselected, nonfunctional protein-protein interactions. I discuss evidence for the existence of these non-functional interactions in kinase-substrate networks from the analysis of the evolution of phosphoproteomes of mammals and fungi.


October 21st, 2009
2:30 to 3:30 pm

"Dominant Pathways in Protein Folding"

Dr. Henri Orland
Institut de Physique Théorique, France

McIntyre Medical Building, Room 903, 3655 Promenade Sir-William-Osler, Montreal, H3G 1Y6 1

Protein folding can be described by the Langevin dynamics. This dynamics can in turn be represented by a “path integral”, which is a weighted sum over all paths joining the denatured state with the native state of the protein. We show how one can compute the dominant paths (paths with largest weight) and how one can calculate dynamical quantities such as rates or transition path times from these paths. The method is illustrated on various simple examples.


October 14, 2009
2:30 to 3:30 pm

"Deriving Executable Models of Biochemical Network Dynamics from Qualitative and Semi-Quantitative Data"

Derek Ruths
McGill University,School of Computer Science

McIntyre Medical Building, Room 903, 3655 Promenade Sir-William-Osler, Montreal, H3G 1Y6 1

Progress in advancing our understanding of biological systems is limited by their sheer complexity, the cost of laboratory materials and equipment, and limitations of current laboratory technology. Computational and mathematical modeling provides ways to address these limitations through hypothesis generation and testing without experimentation - allowing researchers to analyze system structure and dynamics in silico and, then, design lab experiments that yield desired information about phenomena of interest. These models, however, are only as accurate and complete as the data used to build them. Currently most models are constructed from quantitative experimental data. However, since accurate quantitative measurements are hard to obtain and difficult to adapt from literature and online databases, new sources of data for building models need to be explored. In my research, I design methods for building and executing computational models of cellular networks based on qualitative experimental data, which is more abundant, easier to obtain, and reliably reproducible. Such executable models allow for in silico perturbation, simulation, and exploration of biological systems. In this talk, I will present two general strategies for building and executing Petri net-based models of biochemical networks. Both have been successfully used to model and predict the dynamics of signaling networks in normal and cancer cell lines, rivaling the accuracy of existing methods trained on quantitative data.


October 7, 2009
2:30 to 4:00 pm

"Ensemble predictions of RNA and Protein Structures"

Jerome Waldispuhl
McGill University,School of Computer Science

McIntyre Medical Building, Room 903, 3655 Promenade Sir-William-Osler, Montreal, H3G 1Y6 1

In this talk, I will describe my work in the area of computational structural biology. I will describe new ensemble modeling techniques which can analyze and predict an entire landscape of structural and evolutionary solutions, rather than simple single answer optimizations. This philosophy has a broad impact on our understanding of protein and RNA molecules -- Both of which I have applied this approach to and which I will address in this talk. First, I will introduce a new family of algorithms for investigating the folding landscape of transmembrane beta-barrel proteins based only on sequence information, broad investigator knowledge, and a statistical-mechanical approach using the Boltzmann partition function. This provides predictions of all possible structural conformations that might arise in-vivo, along with their relative likelihood of occurrence. Using a parameterizable grammatical model, these algorithms incorporate high-level information, such as membrane thickness, with an energy function based on stacked amino-acid pair statistical potentials to predict ensemble properties, such as the likelihood of two residues pairing in a beta-sheet, or the per-residue X-ray crystal structure B-value. Then, I will present recent algorithmic advances we have made in the techniques of exploration and analysis of RNA sequence/structure maps, an abstract framework which allows us to bridge different aspects of the sequence/structure relationship. In particular, we have successfully applied these techniques to discover deleterious mutations that radically modify the structure in the Hepatitis C virus cis-acting replication element. At a higher level, we provided evidence that the complete sequence of the 3'UTR of the GB RNA virus C has been optimized to preserve the secondary structure of the evolutionarily conserved stem regions from the destabilizing effect of pointwise mutations.

October 2009

MonBUG Seminar
October 8, 2009
6 to 9 pm
"Reconstructing gene networks by epistatic analysis of phenotype and expression data "
Mads Kaern, OISB

Leacock Building, Room 232, McGill University, 855 Sherbrooke St. West, Montreal 1

One the most successful methods to map genetic networks and pathways predates modern genomics by nearly a century. This method, epistatic analysis, traditionally involves observing the phenotypic impact of mutating two different genes individually and in combination. This talk will introduce the basic concepts involved and discuss recent advances emphasizing the inference of transcriptional regulatory networks. In one study, we investigated if conventional epistatic analysis might be used to highlight interactions within the network regulating the transcriptional response of yeast to DNA damage, and developed a method to specifically identify dynamically modulated functional relationships. Somewhat surprisingly, the network derived from phenotypic data has only a modest overlap with that inferred from microarray data. In a second study, we directly compared epistatic analysis based on phenotypic data and marker gene expression. This analysis demonstrates that the two approaches provide complementary information. While conventional analysis correctly infers the order of genes in metabolic pathways, expression-based analysis specifically highlight regulatory hierarchies. Moreover, combining the two methods allows for a nearly complete network reconstruction with a negligible false discovery rate.