|
|
|
|
|
|
T. Przytycka , PhD
Investigator
NCBI, NLM, NIH
Computational Biology Branch
Building 38A 8S812
Bethesda, MD 20894
Tel: 301-402 1723
Fax: 301-480-9241
Email: przytyck@ncbi.nlm.nih.gov
|
|
|
|
|
|
Research
Interests:
- Biological networks and Systems
Biology
- Protein and genome evolution.
- Application of discrete
algorithms and graph theory to molecular and systems biology
- Protein structure and protein
folding.
(see here for a longer description)
|
Group
Members
Postdoctoral
research position in the AlgoCSB group
BMC Bioinformatics
Theoretical Biology
Insights
Algorithms
International
Society for Computational Biology
New York Academy of Sciences
Polish
Mathematical Society (PTM)
Publications
(all)
Book chapters (6)
Publications
in
discipline - oriented publication
servers:
(list
are partially overlapping)
Publications
in PubMed (Computational Biology)
(~25)
Publications
in DBLP server (Computer Science)
(~45)
Publications
in MathSciNet (Math and some CS) (~30)
Ten representative publications
Zotenko E, Mestre J, O'Leary DP, Przytycka TM. Why do hubs in the
yeast protein interaction network tend to be essential: reexamining the
connection between the network topology and essentiality. PLos Comput Biol 2008 Aug 1;4(8):e1000140 Highlighted
in Nature Genetics Review Genetics
Kann MG, Shoemaker BA, Panchenko AR, Przytycka TM. Correlated Evolution of Interacting Proteins:
Looking Behind the Mirror Tree. J Mol Biol.
2008
Zheng J, Rogozin IB, Koonin EV, Przytycka TM. Support for the Coelomata clade of animals from a
rigorous analysis of the pattern of intron conservation Mol Biol Evol. 2007 Nov;24(11):2583-92. (preliminary
version presented at RECOMB Comparative Genomics 2007)
Przytycka T. Stability of characters and construction of
phylogenetic trees. J Comput Biol. 2007 Jun;14(5):539-49
(preliminary version presented at RECOMB 2007)
Guimarães KS,
Jothi R, Zotenko E, Przytycka TM.
Predicting
domain-domain interactions using a parsimony approach. Genome Biol. 2006
;7(11):R104 
Jothi R, Cherukuri PF, Tasneem A,
Przytycka TM. Co-evolutionary
analysis of domains in interacting proteins reveals insights into domain-domain
interactions mediating protein-protein interactions
JMol Biol. 2006 Sep 29;362(4):861-75 ( Faculty 1000 recommended)
Zotenko E,
Guimarães KS, Jothi R, Przytycka TM.
Decomposition of overlapping protein
complexes: a graph theoretical method for analyzing static and dynamic
protein associations. Algorithms
Mol Biol. 2006 Apr 26;1(1):7.  (preliminary version presented at RECOMB Systems Biology 2005)
Jothi R, Kann MG, Przytycka TM. Predicting protein-protein interaction by
searching evolutionary tree automorphism space. Bioinformatics.
2005 Jun;21 Suppl 1:i241-50. (ISMB
2005)
Przytycka T, Aurora R, Rose GD. A protein taxonomy based on secondary structure. Nat
Struct Biol. 1999 Jul;6(7):672-82.
Przytycka
TM Transforming rooted agreement
into unrooted agreement.
J Comput Biol. 1998 (preliminary
version presented at STOC 1997)
Links to
preliminary versions of
selected book
chapters and encyclopedia articles
Protein interaction network based
prediction of domain-domain and
domain-peptide interactions Guimarães KS & Przytycka TM. Protein-protein
interactions and networks Panchenko
and Przytycka (Eds) Springer, 2008
Frach-Przytycka-Thorup Algorithm for
Agreement of three or more trees; Przytycka TM. Encyclopedia of
Algorithms; Kao, Ming Yang (Ed.) Springer (2008).
Graph theoretical
approaches to delineate dynamics of biological processes. Przytycka TM
& Zotenko in Bioinformatics Algorithms: Techniques and
Application; Ion
Mandoiu and Alexander Zelikovsky (Eds). Wiley 2008
Computational
approaches to predict protein-protein and domain-domain interactions.
Jothi R & Przytycka TM in Bioinformatics Algorithms: Techniques and
Application; Ion
Mandoiu and Alexander Zelikovsky (Eds). Wiley 2008
Hidden Markov Models. Przytycka, TM
in Nature
Encyclopedia of the Human Genome Nature Publishing Group (2003).
Our group focuses
on computational modeling and analyzing of biological processes with
emphasizes on hypothesis and theory-driven questions that are enabled by
large-scale data analysis using (but not
limited to) graph theoretical and algorithmic approaches. We frequently look at these processes
through evolutionary lenses with particular focus on evolutionary forces
that shaped biomolecules on sequence, structure, and network levels.
Analysis of biological networks
Relationships between biological
entities are commonly represented as graphs (networks). Examples of such
networks include protein-protein interaction networks, metabolic networks,
genetic networks, domain co-occurrence networks, and various types of neighborhood
networks. Studies of the properties of such networks can provide important
insight into general principles that govern the organization of these
networks as well as the role of their individual components. We are
interested in understand of static and dynamic properties of these network
(Zotenko et al. 2006) as
well as their evolution (Przytycka
and Yi-Kou 2004). We investigate the relation between the topological
priorities of a network at cells responds to perturbation such as gene
knockout (Zotenko
et al. 2008, Nature
Rev. Gen. 2008, collaboration with Dianne O’Leary, UMD),
receptor stimulation (Zielinski et al. submitted, collaboration with Jacek Capla,
NCI) or diseases such as cancer (see below).
Systems Biology approach to study diseases
Complex
diseases, such as cancer, may manifest themselves in differently in
different patients. In particular, cancer may be caused by different abnormalities
of tumor cells. These abnormalities can, however, contribute to
dys-regulating of a common disease-specific pathway. In a collaboration
with Howard
Fine’s group at NCI, NIH including Stefan Wuchty
we plan to utilize genetic variations and expression data in an attempt to
indentify such dys-regulated pathways in Glioma.
In a
collaboration with Michael
Ferdig’s group (Notre Dame) we utilize the diverse type of data
including eQTL analysis, protein interaction and gene expression to study
regulatory mechanism of Plasmodium and
its response to drugs.
In collaboration
with Jacek
Capla (Radiation Oncology Branch NCI, NIH), we study the interplay
between the fundamental for signaling pathways: EGF, IGF, and insulin by a
mixture of experimental and computational approaches.
Predicting
domain and protein interactions
The
analysis of a network of biological relations cannot be separated from
understanding and predicting the relations themselves. Therefore, an
important component of our research is analysis and prediction of
biological relations that define these networks; for example, prediction of
protein-protein and domain-domain interactions. Using parsimony based approach and
the Linear Programming (LP) optimization technique, I devised an effective
method to predict interacting domains. In addition to its high predictive
power, this approach also allowed for some quantification of the bias in
domain interaction data. Other methods developed and applied in my group,
such as co-evolution and
phylogeny based approaches, also contributed to a better understanding of
the evolution of protein and domain interaction. We have developed a new
parsimony based approach to predict domain interaction (Guimaraes et al. 2006, Guimaraes et al. 2008)
as well as empowered earlier methods such as co-evolution/mirrortree method
(Jothi
et al. 2005, Jothi
et al. 2006, Kann
et al. 2007, Kann
et al. 2008) or phylogenetic profile (Jothi et al. 2007). In collaboration with Maricel Kann, UMBC
we investigate the evolutionary constraints acting on interacting domains
(see also next section).
Evolutionary
forces shaping protein and DNA
sequences
Living organisms
are extremely complex and the numerous genome products work together in a
coordinated fashion to provide specific cellular functions. Similarly
writhing individual molecule specific combinations of amino acids or
nucleotides are required to ensure foldability and/or function. To ensure
functioning of such complex systems physically and/or functionally
interacting components must evolve in mutually dependent fashion. The
effect of such evolutionary pressure is observed in correlated evolution of
interacting proteins and domains (Jothi
et al. 2005, Jothi
et al. 2006, Jothi
et al. 2007, Kann
et al. 2007, Kann
et al. 2008), in constraints on amino-acid changes (Przytycka et al.
2008) or codon usage (Huang et al. in preparation). Our collaborators in this area include David Lipman
(codon usage), Maricel
Kann, UMBC, (correlated evolution of interacting proteins), and Anna Panchenko
NCBI, NLM, NIH (correlated evolution within protein structure and its
relation to allosteric interactions).
Phylogenetic
analysis and inferring evolutionary trees
Studying
of evolutionary forces shaping the properties of bio-molecules in the cell
interleaves with studying of past evolutionary events leading to the
variety of species and individual molecules we observe today. Our studies
include evolution of domain architectures (Przytycka
et al. 2005, in collaboration with Dannie Durand, CMU),
stability of phylogenteic characters (Przytycka
2007) and on resolving a controversy regarding the topology of the phylogenetic tree of animals (Zheng et
al. 2007; in collaboration with Eugene
Koonin, NCBI, NIH).
We are also interested in classifying
and clustering of homology relation between proteins (Jothi
et al. 2006) as well as algorithmic issues related to evolutionary tree
construction (Przytycka
2007, Farach et al. 2000, Przytycka
1998, Farach et al. 1995).
Genetic
polymorphism and phenotypic variations
Classical
quantitative trait loci (QTL) mapping provides methods for connecting
genotypes (loci) with organism-level phenotypes (quantitative traits), such
as heritable diseases, height, weight, etc. In collaboration with
Dan Camerini-Otero, NIDDK, NIH,
we currently study the relation between differences in the strength
of recombination hotspots and
sequence polymorphism (Zheng et al. 2008, in preparation).
We are
also working on applying expression polymorphism and eQTL (expression
Quantitative Trait Loci) to understand epistatic interactions and gene
regulation. We are particularly
interested in application of these studies to support our systems approach
to study diseases).
Protein
structure and protein folding
Within
the area of protein structure and protein folding we are particularly
interested in the interplay between local and global sequence information
in determining protein fold.
Continuous
increase in the size of protein structure database puts new demands on the
speed of algorithms for protein structure comparison. Yet, structure comparison is
known to be computationally intense task. This necessitates development of
fast approximate structure comparison algorithms that could be used as a filter before using more
computationally demanding approaches. Out research in this area explores
the use of secondary structures either by their direct sequence-like
alignment (Przytycka
et al. Nat. Str. Bio. 1999, in collaboration with George Rose, JHU and
Jeong et al.
2006 in collaboration with Piotr Berman, PSU). We also
developed we developed a fast
method of tertiary structure comparison based on a footprint (or a profile)
of conformations of triplets of secondary structures (Zotenko et al., 2006, Zotenko et al. 2007;
collaboration with Dianne
O’Leary).
We also
explore topological motifs and putative folding pathways of mostly beta proteins (Przytycka
et al. 2002 , in collaboration with George Rose, JHU and Jeong et al.
2008, to appear, in collaboration with Piotr Berman, PSU).
Click here for sample projects.
|
|
Participation in Scientific
Committees (most recent)
|
- Keystone
Symposium Biomolecular Interaction Networks: Function and Disease,
2010, co-chair
- Pacific
Symposium on Biocomputing PSB 2009 (session co-chair)
- Brazilian
Symposium on Bioinformatics, BSB 2009, (co-chair)
- IEEE
21th International Conference on Advanced Information Networking and
Applications, Niagara Falls, Canada, May 21-23,
2007(PC).
- SIAM
Meeting on Discrete Mathematics: Minisymposium in Computational
Biology, June 2006, Victoria,
Canada,
(invited minisymposim organizer).
- DIMACS
Workshop on Sequence, Structure and Systems Approach to Predict
Protein Function, Rutgers, May 2006,
(co-organizer)
- 2nd International Workshop on Bioinformatics
Research and Applications, University
of Reading, UK,
May 2006, (PC)
- The
First Annual RECOMB Satellite
Workshop on Systems Biology San Diego,
December 2005, (PC and OC)
- CompBioNets
2005 Algorithms and Computational Methods for Biochemical and
Evolutionary Biology; Lyon,
December 2005, (PC)
- DIMACS
Workshop on Biomolecular Networks: Topological Properties and
Evolution Rutgers, May 2005 (co-organizer)
|
|
Recent Invited Conference Talks
|
- 2008
INFORMS Annual Meeting, Oct 12-15, 2008, Washington DC. Invited organizer and speaker of the
session Optimization Problems
in Complex Biological Systems.
- 2008
SIAM Annual Meeting, July 7-11, 2008 in San Diego, California Invited
organizer of a special session Networks: biological, social and
internet
- bi-annual SIAM Conference on Discrete
Mathematics June 16-19, 2008 University of Vermont in Burlington,
Vermont – invited minisymposium speaker.
- 1st
Canadian Discrete and Algorithmic Mathematics Conference, Banff May 2007
(speaker in invited organizer of the minisymposium: Combinatorial
problems in genomics)
- 39th
Symposium on the Interface: Systems Biology Philadelphia, May, 2007,
invited speaker.
- First
Bertinoro Systems Biology Meeting, Bertinoro, Italy,
May 2007, invited speaker.
- SIAM
Meeting on Discrete Mathematics Minisymposium in Computational
Biology, June 2006, Victoria,
Canada.
- 2nd International Workshop on Bioinformatics Research and Applications
2006 (keynote talk)
- Fifth Virtual Conference on Genomics and
Bioinformatics, October 2005.
- Biological Networks: Interaction with Genome
and Developmental Evolution
Bertinoro, May 2005. BCB
2004 Second Bertinoro Computational Biology Meeting.
|
|
Most recent contributed presentations in highly selective conferences
|
- RECOMB 2008 Regulatory
Genomics/SystemsBiology/DREAM3 Dissecting the
dynamics of yeast transcription factors (joint work with
Raja Jothi, S Balaji, Arthur Wuster,
Joshua A. Grochow, Jorg Gsponer, L Aravind and M. Madan Babu)
- RECOMB 2007 Comparatory Genomics.
A
rigorous analysis of the pattern of intron conservation supports the
Coelomata clade of animals, (joint work with Zheng, Igor
Rogozin, and Eugene Koonin).
- WABI 2007 Bringing folding pathways into strand pairing prediction
(joint work with Jieun Jeong and Piotr Berman)
- RECOMB 2006 An important connection between
network motifs and parsimony models.
- RECOMB 2005 Systems Biology Decomposition
of Overlapping Protein Complexes: A Graph Theoretical Method for
Analyzing Static and Dynamic Protein Associations (joint work with
Elena Zotenko, Katia Guimaraes
and Raja Jothi).
- ISMB 2005 Predicting
Protein-Protein Interaction by Searching Evolutionary Tree
Automorphism Space, (joint work with Raja Jothi and Maricel Kann).
- RECOM
|
