Research
Most proteins function by first binding a ligand, such as another protein, peptide, or small molecule. The Polizzi laboratory aims to learn the rules of protein–ligand binding through the lens of de novo protein design. In de novo design, a protein’s structure and sequence are computed from first principles. This approach has seen much recent success for the creation of new protein shapes. However, the design of proteins that bind to specific ligands remains an outstanding challenge, due to the additional degrees of freedom introduced by a ligand and the subtle balance of energetics. In order to design functional proteins, we must first crack the “binding code.”
Our lab aims to tackle this challenge through a tight coupling of computation and experiment. We develop computational design methods that learn from protein structure and test these algorithms by making new ligand-binding proteins in the lab. We focus on designing proteins that bind to small molecules, as these are the most difficult targets for screening or immunization techniques. We build proteins that not only hone our design methods but also act as useful tools for biology, e.g., for metabolite sensing, proteome editing, and genetic-code expansion in mammalian cells. Ultimately, we hope that by cracking the ligand-binding code we open the door to the design of functional proteins not seen in nature.
Expanding the genetic code with protein design
Aminoacyl tRNA synthetases link amino acids with their cognate tRNAs and are thus critical players in the expansion of the genetic code. However, it has proven incredibly difficult to engineer synthetases to recognize polar, non-natural amino acids that mimic common post-translational modifications (PTMs) of proteins. The ability to precisely program PTMs and monitor their effects on cell fate in disease-relevant mammalian cells has been a longstanding goal. Toward this end, our group uses protein design methods to dramatically alter the substrate specificity of tRNA synthetases, focusing on challenging, polar PTM mimics.
Designing new protein-based tools
Biology can be monitored and manipulated by ligand-binding proteins. Until recently, however, the discovery of binding proteins depended on the serendipity of high-throughput experimental screening and animal inoculations. Now, our group has developed a computational method called COMBS that can generate protein sequences from scratch that are able to tightly and specifically bind to challenging targets. COMBS thus enables the rapid generation of proteins that may act as sensors, therapeutics, or delivery vehicles. We hope that by distilling the principles of ligand binding into a computational algorithm, we can make the design of ligand-binding proteins as routine as the design of primers for PCR, thereby eliminating the bottlenecks in monitoring and manipulating biology at the molecular level and, ultimately, paving the way for the design of de novo function.
Learning contextual signatures of molecular recognition
Our group is exploring new deep learning architectures to understand and predict ligand-binding cooperativity. We leverage custom-built databases to train and test models that learn molecular features that give rise to tight and specific binding of ligands.
Our Team
Our group is part of the Cancer Biology department at Dana-Farber Cancer Institute and the Biological Chemistry and Molecular Pharmacology department at Harvard Medical School.
Nick Polizzi, PhD
Assistant Professor
pronouns: he / him / his
Nick received his PhD in Biochemistry from Duke University, where he developed experimental and theoretical models to probe mechanisms of biological charge flow. He then moved to UCSF for his postdoctoral work, where he built new methods to design proteins from scratch that specifically bind to small molecules. At Harvard Medical School and Dana-Farber Cancer Institute, Nick aims to crack the “binding code” by designing ligand-binding proteins that can serve as useful tools for observing and manipulating biology.
NicholasF_Polizzi@DFCI.harvard.edu
Google Scholar
@nickpolizzi_
github
Jody Mou
Graduate Student
Jody graduated from Johns Hopkins University with a degree in biomedical engineering and is currently a PhD student in the Harvard-MIT HST program. During undergrad, she worked on molecular immunology at JHU and machine learning for protein engineering at Microsoft Research. Jody is working on de novo design of ligand-binding proteins using computational and deep learning approaches. She is interested in synthetic biology and research tool development. Outside of the lab, she enjoys making things and exploring new places.
Chun-Chen Jerry Yao
Graduate Student
Jerry is a graduate student in Harvard’s Molecules, Cells, and Organisms (MCO) program. He earned his BSc at MIT in Brain and Cognitive Science (BCS), where he optimized an in situ sequencing-by-synthesis method in Ed Boyden’s lab and Fei Chen’s lab; he also studied transcription-translation uncoupling in Bacillus subtilis in Gene-Wei Li’s lab. During his gap year before starting at Harvard, he served the Taiwanese army and then worked with Dan Tawfik at the Weizmann Institute to investigate the origin of peptides and the genetic code in the origin of life. For his PhD, he hopes to elucidate the principles of biomolecular evolution using protein design, evolutionary biology, and high-throughput experimental approaches. When he is not thinking about science, he enjoys cooking, hiking, backpacking, learning languages, and exploring the evolution of human artifacts.
Jeffrey Chang
Graduate Student
Jeffrey is a graduate student in Harvard's physics department. As an undergraduate, he studied fluorescent protein photochemistry in Prof. Steve Boxer's lab at Stanford. Earlier in graduate school, he worked in statistical mechanics education with Prof. Steve Kivelson, analysis methods for single-molecule biophysics with Prof. Wesley Wong, and high-throughput assay development with Prof. Michael Desai. Currently, he wants to understand the basic principles of biomolecular specificity by designing peptide-binding proteins. In his spare time, he kickboxes, dances Lindy Hop, and plays in a jazz quartet called Mutual Beef.
Jing Sun
Graduate Student
Jing is a graduate student in the Biological and Biomedical Sciences (BBS) Program at Harvard. She graduated from Cornell University with a bachelor degree in Biological Sciences and in Statistical Science. During her undergrad, Jing studied viral proteins using X-ray crystallography at Cornell and worked on an ABC transporter using cryo-EM at Scripps Research. She enjoys skiing, traveling, eating, and playing with cats.
Ben Fry
Graduate Student
Ben graduated from Johns Hopkins University with his Bachelors in Biophysics and a minor in Computer Science. He is currently a PhD student in Harvard’s Biophysics Program. As an intern with the Rosetta Commons, he worked in Dr. Daniel Kulp’s lab at The Wistar Institute to develop a computational protocol that targets viral glycoprotein epitopes for immune recognition. He continued this work at Hopkins under Dr. Jeffrey Gray. He is broadly interested in developing new computational techniques for protein design with the hope that they can be used to create the next generation of protein-based therapeutics. Outside of the lab, Ben enjoys exploring the city, hiking, cooking, and playing guitar.
Kaia Slaw
Research Technician
Kaia graduated from Bucknell University with a BSc in Cell Biology/Biochemistry. During her undergrad, she worked in the Switzer Lab to characterize human DNA methyltransferase with HSAN1E-associated mutations. Currently, she is excited to characterize de novo proteins created through computational methods. Outside of the lab, she enjoys crafting, taking care of her plants, and working out.
Angela Mei
Undergraduate Student
Angela is an undergraduate at Harvard studying computer science and chemistry. She is fascinated by the potential of using computational tools to revolutionize protein design, with a particular interest in how these advancements could lead to therapeutic applications. In her free time, Angela enjoys swimming, learning new languages, and watching sports.
Franzi Sendker, PhD
Postdoctoral Research Fellow
Franzi grew up in Germany and graduated from Albert Ludwigs University Freiburg with a BSc in Molecular Medicine and afterwards a MSc in Biochemistry from University of Leipzig. She pursued her PhD at the Max Planck Institute for Terrestrial Microbiology where she worked with Dr. Georg Hochberg in the area of Evolutionary Biochemistry. In her work she investigated self-assembly of proteins and the evolution of biological complexity. Outside the lab, Franzi enjoys playing Tennis or any other racket sport, cooking, and going to the movies.
Roksana Azad, PhD
Postdoctoral Research Fellow
Roksana grew up in Bangladesh and later immigrated to the United States, where she completed her B.S. with academic honors in Chemistry and Biochemistry from York College. She received her Ph.D. in Biochemistry and Molecular Biophysics from the CUNY Advanced Science Research Center working in Prof. Kevin Gardner's lab. During her Ph.D., Roksana studied the superfamily of PAS/LOV sensory domains and their regulatory control of Ser/Thr kinases in plants and humans. Her thesis research characterized the structure and function of the allosteric regulation of PAS to control the kinase function using integrated techniques including NMR (high-pressure, 2D/3D), HDX-MS, cryo-EM, and X-ray. In her free time, Roksana is passionate about volunteering in STEM outreach, and she enjoys cooking, watching soccer and cricket, outdoor activities, and keeping up with recent scientific discoveries.
Azim Dharani
Graduate Student
Azim is a graduate student in Harvard's Chemical Biology program. He studied Chemistry and Computational Biology as an undergraduate at Duke and obtained a Master's degree in Chemistry from the University of Cambridge, where he developed enzyme-coated nanomaterials for carbon capture. After graduation, Azim worked at the D.E. Shaw Research Institute, using molecular dynamics simulations to inform drug discovery. In the lab, he is interested in leveraging high-throughput screening to improve the prediction of protein-ligand complexes. Outside the lab, he enjoys playing basketball, reading science fiction, and cooking.
Amanda Judge
Executive Support Specialist
Amanda has been with DFCI Cancer Biology since 2016. She enjoys learning history, going to museums, listening to too many podcasts, and snuggling with her grouchy cat. She was a sociology major so she won’t be able to help with science questions but would be very happy to help with anything else.
Lab Alumni
Varun Ullanat Current Position: Master’s Student, Biomedical Informatics, HMS
James Davey Current Position: Assistant Professor, Dalhousie University, Nova Scotia, Canada
Tessa Haining Current Position: Rhodes Scholar, Oxford UK
Marco Hansel Current position: Harvard undergraduate
Gabe Au Current position: Johns Hopkins undergraduate
Kevin Tan Current position: Microsoft
Masy Domecillo Current position: Caltech graduate student
Amina Menhour
Anna Lian Current position: MIT undergraduate
Jesús Valdiviezo Mora, PhD
Rotation students: Ryan Acbay, Bryan Banuelos Jara, Jio Jeong, Fiona Qu, Matt Knotts, Lilia Evgeniou
publications
De novo design of drug-binding proteins with predictable binding energy and specificity.
L. Lu, …, A. Ashworth, N. F. Polizzi, W. F. DeGrado. Science. 384, 106–112 (2024).
PubMed ID: 38187746
Github Repos: combs2
News Coverage: Chemistry World
AF2BIND: Predicting ligand-binding sites using the pair representation of AlphaFold2.
A. Gazizov, A. Lian, C. Goverde, S. Ovchinnikov, N. F. Polizzi. bioRxiv. doi: https://doi.org/10.1101/2023.10.15.562410 (2023).
Github Repos: af2bind
Google Colab: af2bind
A defined structural unit enables de novo design of small-molecule-binding proteins.
N. F. Polizzi, W. F. DeGrado. Science. 369, 1227–1233 (2020).
PubMed ID: 32883865
Github Repos: combs
Perspective in Science by Dr. Anna Peacock
Research Highlight: Nature Methods
News Coverage: Chemistry World, Phys.org
De novo design of a hyperstable non-natural protein-ligand complex with sub-Å accuracy.
N. F. Polizzi, Y. Wu, T. Lemmin, A. M. Maxwell, S.-Q. Zhang, J. Rawson, D. N. Beratan, M. J. Therien, W. F. DeGrado. Nature Chemistry. 9, 1157–1164 (2017).
Allosteric cooperation in a de novo-designed two-domain protein.
F. Pirro, N. Schmidt, J. Lincoff, Z. X. Widel, N. F. Polizzi, L. Liu, M. J. Therien, M. Grabe, M. Chino, A. Lombardi, W. F. DeGrado. Proc. Natl. Acad. Sci. U. S. A. 117, 33246–33253 (2020).
PubMed ID: 33318174
Deposited Structures: 7jh6
Lab news
Aug 13th, 2022
Jody and Jerry officially join the lab! We are thrilled to continue our work together on some of the most pressing issues in ligand-binding protein design!
May 8th, 2022
Jody was awarded the T.S. Lin Fellowship at MIT, which will fund her graduate studies for a year! Congrats, Jody!
May 3rd, 2022
Welcome to Jerry and Jody, who are starting their rotations in the group! Jerry joins from Harvard’s MCO program, and Jody from the joint Harvard-MIT HST program.
Doors open!
The Polizzi Lab opened its doors on May 2, 2022. Our lab is located on the 4th floor of the Longwood Center (LC), one of the newest buildings in the Longwood / Harvard Medical School area. The LC has incredible core facilities and research space for pursuing our protein design work. We’re situated among other Cancer Biology faculty, which makes for easy collaboration!
Twitter feed
Resources
Tutorials
Media
Design of small-molecule-binding proteins
Nick gives a 20 min talk @ Virtual Biocatalysis and Protein Engineering Meetup
Join the team
We consider applications on a rolling basis. If you are interested in joining a growing, collaborative team working at the forefront of protein design, please email Nick your CV and a statement explaining your goals and interests.
Postdocs
We are always looking for highly motivated postdocs with computational and/or experimental skills and a passion for protein structure/function/design. We have lots of exciting projects involving protein design, high-throughput experiments, and structure determination.
Graduate students
We’re recruiting graduate students who are accepted into Harvard or Harvard/MIT PhD programs! Please reach out to Nick directly via email with a description of your interests. (MIT graduate students who are not in a joint program can be co-advised with an MIT faculty.) Prospective graduate students are also encouraged to reach out, although in order to join the group they must first be admitted into a PhD program.
Contact Us
Executive Support:
Amanda Judge: Amanda_Judge@DFCI.harvard.edu
Principal Investigator:
Nick Polizzi: NicholasF_Polizzi@DFCI.harvard.edu
Location
360 Longwood Ave
Longwood Center (LC)
Boston, MA 02215
Office: LC-4312
Lab: LC-4302
