Computer Science Assistant

RESUME - DIMITRIJE JEVREMOVIC

Address **** *** ** **, #***

Minneapolis, MN 55414

United States of America

E-mail ******@**.***.***

http://www.cs.umn.edu/~jevrem

Web page

http://www.linkedin.com/in/jevrem

INTERESTS: computational biology and bioinformatics, parallel and high performance computing,

data mining and machine learning

EDUCATION

Ph. D. Candidate in Computer Science (September 2007 May 2013)

University of Minnesota, Twin Cities, MN

Major: Computer Science (GPA: 3.85)

Minor: Biomedical Informatics and Computational Biology

Thesis: Scalable Computation and Analysis of Elementary Flux Modes in Metabolic Networks

M. S. in Computer Science (September 2007 January 2010)

University of Minnesota, Twin Cities, MN

Graduate coursework: (GPA: 3.85)

Applications: Data Mining, Advanced Database Systems, Data Mining in Bioinformatics

Theory: Advanced Data Structures and Algorithms, Machine Learning, Computational

Matrix Theory

Bio: Statistics in Human Genetics and Biology, Molecular Cell Biology, Microbial Genomics

& Bioinformatics, Computational Techniques in Genomics, Systems Analysis of Biological

Processes

Systems: Advanced Internet Programming, Parallel Computing, Data Comm. and Comp.

Networks

M. S. in Computer Science and Informatics (October 2005 July 2007)

Faculty of Technical Sciences, University of Novi Sad, Serbia

Major: Computer Science

(transferred to Univ. of Minnesota in 09/2007)

B. S. in Computer Science and Informatics (October 2000 July 2005)

Faculty of Technical Sciences, University of Novi Sad, Serbia

Major: Computer Science (GPA: 3.75)

Thesis: Environment for the development of persistent elements of object data models

WORKING EXPERIENCE

IBM Ph.D. Student Fellow, Graduate Research Assistant (September 2010 May 2011,

September 2011 May 2012, September 2012 May 2013)

University of Minnesota, Department of Computer Science

Doctoral research in the parallel computation and analysis of metabolic pathways in

metabolic networks under quasi steady state conditions

Page 1 of 4

Implementation of parallel algorithms using MPI/OpenMP libraries and partitioned global

address space libraries (e.g. Global Arrays)

Design of algorithms for enumeration of subsets of reaction/gene knockouts to optimize the

production of chemical in modified microorganism strains based on elementary flux modes

Use of linear and non-linear optimization methods to infer favorable genetic modifications

in the cellular metabolic networks

Developments maintained and available at http://elmocomp.sourceforge.net

Bioinformatics Intern (June 2012 September 2012)

Genentech, Bioinformatics Department, South San Francisco, CA

design of software pipeline for (1) efficient computational prediction of 3D protein structure

and protein structural features (2) mapping of somatic human genome mutations onto

protein structure (3) family-specific spatial clustering of mapped somatic mutations for the

discovery of mutation hotspots

tools used: MODELLER, HH-Suite, BLAST, PyMol; software development in (Bio)Python;

data repositories used: PDB, UniprotKB/Swiss-Prot, COSMIC, SCOP/CATH

Summer Research Assistant (June 2011 August 2011)

NEC Labs America @ Machine Learning Group, Princeton, NJ

Study of the metabolic network models of the green microalgae Chlamydomonas reinhardtii

and strategies for the finding of genetic modifications required to create efficient lipid

producing strains

Utilization of the optimization-based methods (OptKnock, RobustKnock, GDLS, OptGene)

and COBRA framework to enumerate efficient gene/reaction modification targets

Literature review and technology scouting

Computational Biology Intern (June 2010 August 2010)

Dow AgroSciences, Indianapolis, IN

Implemented software pipeline for the alignment of short-read DNA sequences, obtained by

using second-generation sequencing instruments, to the reference genome (PipelinePilot,

PilotScript, Perl, shell scripting).

Conducted literature reviews for the purpose of in-house implementation of in vivo protein

structure discovery and epigenomics capabilities

Graduate Research Assistant and Trainee (September 2007 May 2010)

University of Minnesota, Department of Computer Science

Development of the algorithm for serial and parallel computation of metabolic pathways in

metabolic network

Parallel algorithm development in C++/STL/MPI/OpenMP for different computing platforms

(Blue Gene/P; SGI Altix XE machines).

Elementary mode analysis; Flux balance analysis; metabolic network modeling

Implementation of various bioinformatics algorithms (sequence analysis, microarrays)

Graduate Research and Teaching Assistant (October 2005 July 2007)

Faculty of Technical Sciences, University of Novi Sad, Serbia

Developed drawing editor application for the simulation of basketball actions as a software

tool used by basketball coaches, managers, and professionals (C/C++ Borland

Builder/VCL/UML/Power Designer/OpenCV)

Taught lab classes in C/C++/relational DB modeling. Organized examinations and graded

Page 2 of 4

students work

Summer Intern (July 2004 September 2004)

M-otion GmbH, Vienna, Austria

Documented corporate application server and scripting language.

Summer Intern (July 2002 September 2002)

Provincial Council of Alicante, Alicante, Spain

Web application development in ASP, MS SQL Server and Java Script

TECHNICAL SKILLS

Programming Languages and Libraries:

C and C++ (with STL)

Java/J2EE, HTML/XML/XSL, UML, SQL

Scientific computing: Matlab, Octave, R (Bioconductor)

Scripting: PHP, Ruby, JavaScript with AJAX, Perl and CGI, Python

Parallel computing : MPI, OpenMP, pthreads, Global Array Toolkit, CUDA, Hadoop

Other: Subversion

Bioinformatics Tools:

Bioconductor, BioPerl, ProteinPilot, Spotfire, Expressionist, PipelinePilot

Metabolic network analysis: COBRA, SBML, Elementary flux mode analysis

Sequence analysis: BLAST, GeneMark, Glimmer, ClustalW,

Databases: PDB, EMBOSS, GenBank, SWISS-Prot, Clover

Microarray analysis: SAM, Cluster and TreeView

Development Frameworks and Tools:

Eclipse, Borland Builder C++ 5/6, IBM XL compiler suite, Intel C++ compiler, GNU

gcc and g++

Database modeling and systems: MS SQL Server, MySQL, PowerDesigner

Web development: JBoss, Apache Tomcat, Apache HTTPD server, Hibernate, Struts

Data Mining Tools: Weka

Operating Systems: Windows, Unix/Linux

LANGUAGES

English (fluent)

Spanish (fluent)

Serbian (native)

German (intermediate)

TRAINING

Scaling to petascale summer school, Virtual School of Computational Science and

Engineering, Univ. of Ill. Urbana-Champaign, August 2009.

Many-core processors summer school, Virtual School of Computational Science and

Engineering, Univ. of Ill. Chicago, August 2009.

AWARDS AND HONORS

Page 3 of 4

IBM Ph.D. Fellowship Award for 2010/2011, 2011/2012 and 2012/13

Biomedical Informatics and Computational Biology fellowship (2007-2009), University of

Minnesota Rochester

Technical Committee on Parallel Processing travel grant to attend IPDPS 2011 conference in

Anchorage, AK

Harden Bursary to attend the Metabolic Pathways Analysis 2011 conference in Chester,

England

INVITED TALKS

Scalable Computation and Analysis of Elementary Modes in Metabolic Networks,

Metabolic Pathway Analysis, 71st Harden Conference, September 2011, Chester, England,

(talk + poster session)

"Computation and Analysis of Metabolic Pathways", Biomedical Informatics and

Computational Biology semi-annual symposium, Jan. 2011

PUBLICATIONS

D. Jevremovic, D. Boley, Parallel Computation of Elementary Flux Modes in Metabolic

Networks using Global Array Toolkit, The 6th Conference on Partitioned Global Address

Space Programming Models, Santa Barbara, CA, 2012

D. Jevremovic, D. Boley, Finding minimal generating set for metabolic network with

reversible reactions, submitted for review, 2012

D. Jevremovic, D. Boley, C. Sosa, Algorithms for enumeration of efficient reaction

knockouts using elementary flux modes, submitted for review, 2012

D. Jevremovic, D. Boley, C. Sosa, Divide-and-conquer approach to the parallel

computation of elementary flux modes in metabolic networks, 2011 IEEE International

Symposium on Parallel and Distributed Processing Workshops and PhD Forum, pp. 502-511

D. Jevremovic, C. Trinh, F. Srienc, C. Sosa, D. Boley, Parallelization of the Nullspace

Algorithm for the Computation of Metabolic Pathways, Parallel Computing, vol. 37, no. 6 -

7, (2011), pp. 261-278

D. Jevremovic, C. Trinh, F. Srienc, D. Boley, On Algebraic Properties of Extreme Pathways

in Metabolic Networks, Journal of Computational Biology, vol. 17, no. 2, (2010), pp. 107-119

D. Jevremovic, C. Trinh, F. Srienc, D. Boley, A Simple Rank Test to Distinguish Extreme

Pathways from Elementary Modes in Metabolic Networks, technical report 08-029, 2008

Page 4 of 4

Contact this candidate