Rules of Protein-DNA Recognition: Computational and Experimental Advances (18w5045)

Arriving in Oaxaca, Mexico Sunday, June 3 and departing Friday June 8, 2018


Marcus Noyes (New York University, Langone Medical Center)

(Boston University)


This workshop will focus on recent progress in measuring and modeling the biophysical mechanisms that enable DNA-binding proteins (such as transcription factors) to bind their cognate sites within the vast genome. The main purpose of the workshop is to bring together researchers with diverse mathematical, computational and experimental approaches to studying protein-DNA recognition. The goal will be to obtain new mathematical and computational approaches to transcription factor binding measurement modeling and prediction. This field has developed from pioneering work on sequence analysis by researchers such as Michael Waterman and Temple Smith, and mathematical representations of protein-DNA binding specificity by Peter von Hippel and Gary Stormo (confirmed participant and previous organizer). A recent successful workshop on algorithmic challenges in genomics held at the Simons Institute for the Theory of Computing (UC Berkeley), organized by Martin Vingron (confirmed BIRS participant) and attended by a number of our confirmed participants, outlined a number of mathematical and computational issues being addressed by the field. We anticipate that the BIRS conference would function to integrate these issues with the current experimental techniques and data, to directly address possible routes forward. A second goal is to improve methods for detecting the impact of genomic DNA differences between people (e.g., single nucleotide polymorphisms, SNPs) on transcription factor binding and gene regulation. Advances in modeling the impact of genomic variation on gene regulation will have tremendous impact on understanding the link between genomic variation and human health and disease. Recent breakthroughs in developing high-throughput experimental techniques for examining the influence of non-coding SNPs on transcription factor binding and function make the development of novel mathematical and computational frameworks for analyzing these datasets a high priority. The proposed workshop, if approved, will help fulfill this urgent need.

The organizer, Dr. Trevor Siggers, is an Assistant Professor in the Department of Biology at Boston University. He has been instrumental in developing and applying the protein-binding microarray (PBM) technique for the high-throughput characterization of protein-DNA binding specificity. His research integrates computational models of protein-DNA binding with genome-scale experimental datasets to understand specificity of gene-regulatory networks and transcription factor evolution. He was awarded a National Science Foundation (NSF) Postdoctoral Research Fellowship in Biological Informatics, a National Institutes of Health (NIH) K22 Research Scholar Development award, and an NIH R01 that utilizes computational and experimental approaches to study proteins coordinating inflammation. His laboratory is currently funded by the NSF and NIH.

The co-organizer, Dr. Marcus Noyes is an Assistant Professor in the Department of Biochemistry and Molecular Pharmacology and a core faculty member of the Institute for Systems Genetics at the NYU School of Medicine. Dr. Noyes developed a bacterial one-hybrid assay for the simple characterization of transcription factor specificity, a method that has been employed by a variety of labs to characterize factors from a host of model organisms. In fact, over half of the transcription factors in the fruit fly have been characterized by this method. For this work he received the Harold M. Weintraub Award and was recruited to start his own lab directly out of graduate school as a Lewis-Sigler Fellow at Princeton University. His work at Princeton demonstrated that screens of synthetic transcription factors can be used to accurately predict the function of naturally occurring factors. He has recently accepted a position at NYU where he will continue to these works.

Our invitation to participate in the workshop has been met with great enthusiasm by the researchers in the field. At this point, 29 principal investigators have confirmed their interest in attending the workshop if it is approved (please see the detail list below). The confirmed participants include prominent experimental biophysicists and biologist who are the forefront of the field, as well as mathematicians and physicists whose main focus is on mathematical modeling of protein-DNA interactions. We have taken special care to include a significant number of promising young scientists in the list (e.g., Phil Bradley, Remo Rohs, Raluca Gordan, Matthew Slattery, Matt Weirauch, Polly Fordyce). Seven of the confirmed participants are women (Polly Fordyce, Martha Bulyk, Raluca Gordan, Yael Mandel-Gutfreund, Mona Singh, Zhiping Weng, Marianne Walhout). Finally, to promote new perspectives and collaborations, we have invited and have received confirmations from 10 new participants that did not participate in the previous meeting. We are reserving spots for graduate students and postdoctoral associates who are relatively new to the field, as well as for researchers who have relevant “late-breaking" findings. We will be especially careful to include participants from groups traditionally under-represented in the sciences.

One of the main goals for this meeting is to facilitate free exchange of ideas and foster new collaborations, both formal and informal. We believe that BIRS, with its common areas and a beautiful location, will provide an ideal setting for these activities. Similar workshops held in 2013 (Banff) and 2015 (Oaxaca) were hugely successful and the overwhelming consensus has been to continue this dialog every two years. This meeting brings together an unusual group of people who all share a passion for understanding how proteins recognize their DNA binding sites, and fills an important niche that does not overlap with any other conferences. A brief survey of past participants revealed a number of collaborations (~20) that were initiated as a direct result of the two previous BIRS meetings. One resulting publication, between four previous attendees, even acknowledges BIRS as making the joint research possible (Yang et al. (2013) NAR 42:D148). Given the success of these two previous meetings the overall agenda and goals for the proposed 2017 meeting will follow closely the format of these previous meetings. Rather than presenting experiments and modeling in separate sessions, we will mix experimental and theoretical presentations within a given lecture block whenever possible. Because many invited attendees have both experimental and theoretical/computational components in their labs, they will be encouraged to cover recent advances using both approaches in their talks. We will stress the desire for informal presentations and discussions of unpublished data, to maximize the immediate influence that the meeting has on how the field will evolve.

For the structured part of the meeting, we will follow the format in which 30-minute talks (+10 min for questions) will be scheduled in groups of three or four. Every invited participant will be encouraged to give a talk, though it will not be mandatory. Each group of talks will be followed by a tea/coffee break during which the participants will have a chance to discuss the latest lectures in more detail. Interspersing short sessions with breaks provides us the flexibility to allow extended time for discussions should specific topics generate interest; we found this to be particularly beneficial for the 2015 meeting. We will also be sure to schedule social activities such as dinners and hiking trips to promote scientific discussions and forging of new relationships.