ISMB'98 Schedule
Saturday, JUNE 27
18:00 - 22:00 Ontologies for MB tutorial
Sunday, JUNE 28
8:30 - 12:30 Tutorials (including coffee break)
McClure: Molecular Phylogenetics (4h)
Guex, Peitsch: Comparative protein modelling (4h)
Baldi, Brunak: Bioinformatics: The Machine Learning Approach (4h)
Brazma, Jonassen: Sequence Pattern Discovery Methods (3.5h)
Bueschking, Schleiermacher: WWW-Based Sequence Analysis (3.5h)
12:30 - 13:30 Lunch (for those who registered to at least one tutorial)
13:30 - 17:30 Tutorials (including coffee break)
Pearson: Protein Evolution (4h)
Sankoff: Comparative Genomics (3h)
Lawrence: Bayesian Inference Algorithms (4h)
Baldi: Hidden Markov Models (3h)
Brusic, Zeleznikow: Knowledge Discovery and Data Mining (4h)
18:00 - 20:00 Welcome reception
Monday, JUNE 29
8:00 - 8:30 Coffee/muffins/fruits
8:30 - 9:00 Welcome and announcements
9:00 - 10:00 Invited Speaker Session (Chair J. Glasgow)
Shoshana Wodak
Database Derived Potentials For Prediction of Protein Structure And Stability
Understanding how the amino-acid sequence of proteins determines their
3D structure, and how both sequence and structure determine function,
is of far reaching fundamental importance in molecular biology, and
its many applications, ranging from genomics, to drug and protein
design, disease control, and many other areas. Substantial progress
has been achieved over the last decade in describing the factors that
govern protein stability and in unraveling the mechanism of protein
folding. This has renewed the interest in developing methods for
predicting the 3D structure of protein from their amino acid sequence
and for simulating protein folding or unfolding in the computer. To
perform these tasks reliably the methods must embody at least some, if
not all, of the key features that underlie the true physical
phenomena. This clearly needs the links between energetics and
structure to be established. Such links are typically provided by
molecular mechanics force-fields, which are firmly based on the
principles of physics. But such force-fields can presently not be used
to fold proteins on the computer, because they require a detailed
atomic representation of the system that involves an astronomical
number of degrees of freedom and present day computers are unable to
explore those efficiently enough. Devising force-fields in terms of
reduced descriptions of the protein conformation has therefore been a
recurring theme. It received new impetus recently with the realization
that the body of protein sequences and structure data has probably
reached a sufficient size to derive from it 'effective potentials',
which provide an intermediate description between those given by
detailed atomic force-fields on the one hand, and single residue
specific secondary structure propensities on the other. In recent
years an impressive number of studies has been devoted to this issue.
Here we will present a brief overview of these developments with
emphasis on potentials derived from statistical analyses of known
protein structures. We will start by examining some of the current
thinking about the links between protein energetics and structure, in
order to define the key issues that the derived potentials should be
able to address. To follow, some of the most common types of
potentials will be described and their performance will be illustrated
using examples from work performed in our laboratory. These examples
will include the structure prediction of short peptides and early
folding regions in proteins, fold recognition procedures using
sequence-structure screening methods, and the evaluation of changes in
protein stability caused by mutations. Factors that may limit the
performance of the potentials and perspectives for future development
will be discussed.
10:00 - 10:30 Coffee Break
10:30 - 12:30 Session 1 - Structure and Folding (Chair J. Glasgow)
Genetic Algorithms for Protein Threading
J. Yadgari, A. Amir, R. Unger
Hierarchical Minimization with Distance and Angle Constraints
J.R. Gunn
Modeling Protein Homopolymeric Repeats: Possible Poly Glutamine
Structural Motifs for Huntington's Disease
R.H. Lathrop, M. Casale, D.J. Tobias, J.L. Marsh, L.M. Thompson
A Surface Measure for Probabilistic Structural Computations
J.P. Schmidt, C.C. Chen, J.L. Cooper and R.B. Altman
12:30 - 14:00 Lunch
14:00 - 16:00 Session 2 - Hidden Markov Models (Chair F. Major)
A Hidden Markov Model for Predicting Transmembrane Helices
in Protein Sequences
E.L.L. Sonnhammer, G. von Heijne, A. Krogh
Prediction of Signal Peptides and Signal Anchors by
a Hidden Markov Model
H. Nielsen and A. Krogh
Identification of Divergent Functions in Homologous
Proteins by Induction over Conserved Modules
I. Shah, L. Hunter
Computational Applications of DNA Physical Scales
P. Baldi, S. Brunak, Y, Chauvin, A.G. Pedersen
16:30 - 18:00 Poster and Demo Session/Reception
Tuesday, JUNE 30
8:30 - 9:00 Coffee/muffins/fruits
9:00 - 10:00 Invited Talk Session (Chair R. Lathrop)
Michael Waterman
Constructing Restriction Maps
The now-classical double-digest approach for mapping a
region of DNA is a source of interesting mathematical and statistical
problems. The problem is NP-Hard even if there are no measurement
errors. Never-the-less by 1987 the entire genome of {\it E. coli} was
mapped with eight enzymes. A fundamentally new molecular biology
approach to constructing restriction maps, {\em Optical Mapping}, has
been developed by Schwartz et al. (1993), which can rapidly produce
ordered restriction maps of single DNA molecules by fluorescence
microscopy. This approach to restriction mapping also suggests
interesting computational problems. However, it is difficult to
estimate directly the restriction site locations of single DNA
molecules based on these optical mapping data because of the precision
of length measurements and the unknown number of true restriction
sites in the data. Thus this approach to restriction mapping also
suggests interesting computational problems. We discuss the use of a
hierarchical Bayes model based on a mixture model with normals and
random noise. In this model we explicitly consider the missing
observation structure of the data, such as the orientations of
molecules, the allocations of cutting sites to restriction sites, and
the indicator variables of whether observed cut sites are true or
false. Because of the complexity of the model, the large number of
missing data, and the unknown number of restriction sites, we use
Reversible-Jump Markov Chain Monte Carlo (MCMC) to estimate the number
and the locations of the restriction sites. Since there exists a high
multimodality due to unknown orientations of molecules, we also use a
combination of our MCMC approach and the flipping algorithm suggested
by our earlier E-M maximum likelihood methods. The study is highly
computer-intensive and the development of an efficient algorithm is
required.
10:00 - 10:30 Coffee break
10:30 - 12:30 Session 3 Bioinformatic Systems (Chair R. Lathrop)
The LabFlow System for Workflow Management in Large Scale
Biology Research Laboratories
N. Goodman, S. Rozen, L.D. Stein
BioSim - A New Qualitative Simulation Environment for Molecular
Biology
K.R. Heidtke, S. Schulze-Kremer
TAMBIS: Transparent Access to Multiple Bioinformatics
Information Sources. An Overview
P.G. Baker, A. Brass, S. Bechhofer, C. Goble, N. Paton,
R. Stevens
A Computational System for Modelling Flexible
Protein-protein and Protein-DNA Docking
M.J.E. Sternberg, P. Aloy, H.A. Gabb, R.M. Jackson, G. Moont,
E. Querol, F.X. Aviles
12:30 - 14:00 Lunch (on your own)
14:00 - 15:30 Session 4 Sequence Alignment and Analysis (Chair C. Sensen)
Sequence Assembly Validation by Multiple Restriction
Digest Fragment Coverage Analysis
E.C. Rouchka, D.J. States
A Map of the Proteins Space - An automatic hierarchical
classification of all known proteins
G. Yona, N. Linial, N. Tishby, M. Linial
A Statistical Theory of Sequence Alignment with Gaps
D. Drasdo, T. Hwa and M. Lassig
15:30 - 16:00 Coffee break
16:00 - 17:30 Session 5 Biological Databases (Chair T. Gaasterland)
Advanced Query Mechanisms for Biological Databases
I-M. A. Chen, A.S. Kosky, V.M. Markowitz, E. Szeto,
T. Topaloglou
IMGT/LIGM-DB: A Systematized Approach for ImMunoGeneTics
Database Coherence and Data Distribution Improvement
V. Giudicelli, D. Chaume, M-P. Lefranc
Automated clustering and assembly of large EST collections
D.P. Yee, D. Conklin
18:30 - Boat Cruise/Banquet
(Presentation of Best Paper and Poster Presentation Awards)
M. Peitsch
Wednesday, JULY 1
8:30 - 9:00 Coffee/muffins/fruits
9:00 - 10:00 Invited Talk Session (Chair D. Sankoff)
Robert Cedergren
Fishing for Function in RNA Form and Features
The recognition, definition and assignment of RNA structure and
function have been enigmatic in genomic data, since 1) there is no
method to identify RNA genes, 2) the RNA four-letter alphabet renders
the prediction of RNA structures from sequences highly degenerate and
3) the three-dimensional structure and function of few RNA molecules
and/or families have been fully characterized. Using RNAMOT, a search
engine developed in our laboratory to integrate secondary and tertiary
structural information and other tools, we have scanned the GenBank
database to ascertain the distribution of some well-known RNA
functional motifs. Among these, are those involved in protein binding
(Tat and Rev), those having a chemical activity (the catalytic
hammerhead and leadzyme motif, the UV-loop) and aptamers binding small
molecules (amino glycosides, etc). All motifs have occurrences
approaching their probability of existence in random sequences
suggesting that the origin of RNA motifs in the database is due to
evolutionary drift of sequences. Occasional, unusual distributions
may reflect the opportunistic use of the motif for a particular
function in a new context. Such may be the case where viruses
associated with the etiology of HIV have a Tat-binding in their RNA
which could provide a means of communication between these viruses.
Whether a given RNA motif can have function within the context of
database (and out of the known context) has been evaluated in the case
of a hammerhead motif found in the repetitive DNA of Schistosomes.
Here, we show that indeed this hammerhead motif is active in vitro and
in vivo. Of particular interest is the fact that the candidate,
substrate gene for this catalytic motif is potentially a critical gene
whose modulation by the ribozyme could have major effects on these
cells. (supported by Natural Science and Engineering Research Council
of CANADA).
10:00 - 10:30 Coffee break
10:30 - 12:30 Session 6 Sequence Alignment and Analysis (Chair T. Littlejohn)
Bayesian Protein Family Classifier
K. Qu. L.A. McCue, C.E. Lawrence
Compression of Strings with Approximate Repeats
L. Allison, T. Edgoose, T.I. Dix
Segment-based Scores for Pairwise and Multiple Sequence
Alignments
B. Morgenstern, W.R. Atchley, K. Hahn, A. Dress
Calculating the Exact Probability of Language-like
Patterns in Biomolecular Sequences
K. Atteson
12:30 - 14:00 Lunch (on your own)
14:00 - 15:30 Session 7 Translation and Evolution (Chair M. Pietsch)
Genexpress: A Computer System for Description, ANalysis and
Recognition of Regulatory Sequences in Eukaryotic Genome
N.A. Kolchanov, M.P. Ponomarenko, A.E. Kel, Yu.V. Kondrakhin,
A.S. Frolov, F.A. Kolpakov, O.V. Kel, E.A. Ananko,
E.V. Ignatieva, O.A. Podkolodnaya, I.L. Stepanenko,
T.I. Merkulova, V.N. Babenko, D.G. Vorobiev,
S.V. Lavyushev, Yu.V. Ponomarenko, A.V. Kochetov,
G.V. Kolesov, N.L. Podkolodny, L. Milanesi,
E. Wingender, T. Heinemeyer, V.V. Solovyev
Phylogenetic Inference in Protein Superfamilies: Analysis
of SH2 Domains
K. Sjolander
The Ribosome Scanning Model for Translation Initiation:
Implications for Gene Prediction and Full-Length cDNA
Detection
P. Agarwal and V. Bafna
15:30 - Closing Remarks
--- July 1 to 12 ---
Festival International de Jazz de Montreal
** Best paper contest