Prospective Graduate Students (Updated Dec 2020)

One funded Graduate Research Assistantship position in open at the Informatics lab at Lehigh University. Our research aims to leverage machine learning, computational geometry, structural biology, parallel computing and computational statistics to identify and explain mechanisms of protein binding specificity. The position is supported by NIH funding.

A second Research Assistantship position in Computational Materials Science is being considered, but has not been formalized yet. This position will leverage machine learning, computational geometry and parallel computing. Please contact Dr. Chen for more information and updates.

Successful candidates are expected to be highly committed and ambitious, looking constantly for original ways to advance their careers through their publications, presentations and achievements. Personal interest in biomedical problems and a desire to find connections to computational research and solutions is essential. Students seeking academic positions after graduation are especially encouraged to apply.

Professor Brian Chen, the Director of the Informatics Lab, works closely with students in all aspects of research, from conception, to design, experimentation, publication and presentation. In addition to learning how to perform multidisciplinary research in bioinformatics, graduate students in the lab will have opportunities to travel to present their findings, to learn how to mentor undergraduate researchers, to learn how to pursue funding, and to receive direct feedback on their writing and presentations.

Details on specific projects and ideas for new projects will be discussed during a phone or on-site visit (if practical). Candidates should contact Brian Chen directly for more information (telephone: (610) 758-4085 email: "Brian's last name" cselehighedu). A CV will greatly assist initial discussions, but all prospective students must ultimately apply to the Department of Computer Science and Engineering


Prospective Undergraduate Researchers

Undergraduate research in bioinformatics is an excellent way to get exposure to careers in academia and biotechnology. Research is also a great way build your resume of accomplishments and positively differentiate yourself in today's historically difficult job market. Neither a long record of courses in biology, computer science, or statistics, nor junior/senior status is necessary to get involved. I have even worked with freshmen (rising sophomores) in the past, very successfully (see my experiences with Drew Bryant and Brad Dodson below). The key to success in undergraduate research is not factual knowledge - which anyone can assemble - but rather motivation and curiosity.

As research developments are always fluid, specific projects will be discussed when you first contact Brian Chen for more information (telephone: (610) 758-4085 email: "my last name" cselehighedu). A transcript and a resume will be important for finding the project that best fits each applicant.


Students I have Mentored

I have been very fortunate to have mentored several undergraduates during my graduate work at Rice University. Our work was very productive, leading to many publications, engineering awards, poster awards, conference travel fellowships, and summer research fellowships. My trainees had majors in Bioengineering, Computer Science, Evolutionary Biology, Electrical Engineering, and Mathematics, and much of their work has led directly to scientific publications.


Drew H. Bryant

Drew H. Bryant

Drew and I worked together from the summer of 2004, when he was a freshman, to approximately mid-2010. Though he was not an experienced programmer when we started, Drew became critical in prototyping a distributed version of Match Augmentation, which ultimately became Geometric Sieving. We have since used this tool for automatically improving motifs using purely geometric and chemical analysis, publishing our work at RECOMB 2006. Drew and I have also considered ways to integrate additional aspects of protein geometry into our motifs, making for matching criteria with significantly improved positive predictive power. Drew is currently a Ph.D. student at the department of Computer Science at Rice University and has since produced an excellent first-author paperand won several awards.


Joseph H. Bylund

Joseph H. Bylund

Joe and I worked together from summer 2005 until spring 2006, studying the problem of representing related functional sites with a single structure. His work led to the development of Composite Motifs that better represent a family of functional sites than simply picking an individual, which is the policy in existing work. In addition, composite motifs avoid an increase in similarity with functionally unrelated proteins, creating a superior representation of related active sites based on multiple structures. Joe is currently a Ph.D. student in the Integrated Program in Cellular, Molecular and Biomedical Studies at Columbia University Medical Center.


Anne E. Christian

Anne E. Christian

Annie and I collaborated from the summer of 2002 until her graduation in 2004. Annie's research set a foundation for many of our current results by helping us to set the basic thresholds of our code. Supported by the CRA-W Distributed Mentor Project, Annie kept a research journal as she ran exhaustive comparisons of runtime and accuracy, identifying standard parameter settings that we still use now. Annie and I also studied the connection between evolutionary conservation and the conservation of protein structure. Having recently completed her MBA at Harvard Business School, Annie is now with Bain & Company.


Amanda E. Cruess

Amanda E. Cruess

Amanda and I have worked together during 2005, where she helped design a more general version of distributed Match Augmentation which supports significantly more flexible input, as well as greater computational efficiency. Over the summer of 2005, Amanda was supported by the CRA-W Distributed Mentor Project, and we spent the summer exploring the design of motifs using a wider range of structural information. In addition, Amanda has done excellent work in the development of a highly detailed dataset for testing these new motifs, which will appear in upcoming publications. Amanda is currently a software developer at National Instruments.


Anand P. Dharan

Anand P. Dharan

Anand and I worked together during the summer of 2004, while he was participating in the W.M. Keck Center's Undergraduate Research Training Program (URTP). Anand and I studied applications of nonparametric kernel density estimation using Gaussian kernels, for the purpose of measuring the statistical significance of matches. This work formed a prototype for the statistical analysis used in Geometric Sieving, presented at RECOMB 2006, and won first place in the Keck URTP Research Symposium Poster Contest in 2004. After completing his MBA at Harvard Business School, Anand is now an associate with McKinsey and Company.


Brad D. Dodson

Brad D. Dodson

I worked with Brad, a Rice University Century Scholar, from spring 2005 until summer 2006. As a freshman with little biological training, Brad quickly helped develop an interesting statistical analysis, based on Very Large DataBase (VLDB) techniques, for accelerating Geometric Sieving. This technique uses confidence bounds for estimating the median of a distribution of similar active sites, in proteins. Coding analyses of these confidence bounds, we were able to accelerate Geometric Sieving performance nearly 10 times over our original sampling technique. This work was presented at RECOMB 2006. Brad and I also investigated the mathematical properties of matching pointsets. Brad is currently a software development engineer at Microsoft.


Jessica Wu

Jessica Y. Wu

Jessica and I worked together during the spring of 2005 on generalizing the Geometric Sieving technique. Working together with Amanda E. Cruess and Brad D. Dodson, Jessica designed a very flexible parser for accepting a generalized experiment-design file format. This foundational work has permitted us to very efficiently pursue complex investigations using Geometric Sieving and a large supercomputing cluster. Jessica is now collaborating with Amarda Shehu on Kinematics and Conformational Issues for Molecules. Jessica is currently a Ph.D. student at the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology.