Research

Engineering protein agonists, antagonists, and enzymes

Directed protein evolution adapts Darwinian principles of natural selection and applies them to individual molecules in the laboratory to rapidly identify the 'fittest' members in a pool of protein sequences. Employing tools from molecular biology, we construct protein libraries > 1013 in size and use these libraries to interrogate protein sequence-function relationships and to engineer new therapeutics for the treatment of cancer, diabetes, and infectious diseases. An important part of our research is the development of new in vitro and in vivo directed evolution platforms to facilitate selections for protein properties that are more complex than binding affinity.

Collaborators:
Ravi Radhakrishnan, Penn; Bill DeGrado, Penn; Erica Ollmann Saphire, Scripps; Rex Ahima, Penn; Barry Cooperman, Penn.



Understanding cellular decision making in lineage commitment

Although it is well appreciated that stem and progenitor cells make discrete fate choices in response to their environment, it is less clear how these all-or-none decisions are reached in the presence of a smooth gradient of a single stimulus or in an environment containing multiple conflicting cues. We are developing mathematical models that integrate cell-extrinsic and cell-intrinsic cues in order to better understand how these factors influence commitment decisions. We are also experimentally testing our model-driven hypotheses in bipotent and multipotent progenitor cells to identify and characterize novel regulatory mechanisms that can confer robustness to a cell's decision-making process.

Collaborators:
Mitch Weiss, CHOP; Chris Chen, Penn; Kurt Hankenson, Penn; Jason Burdick, Penn; Rob Mauck, Penn; Seth Corey, Northwestern.



Network analysis and cellular programming through synthetic biology

Biomolecular networks are difficult to study in their native context due to the multitude of confounding factors present in this environment. In our synthetic biology approach, we transfer a biomolecular network into an artificial host and apply a variety of perturbations in order to understand how the system operates in this isolated environment. Furthermore, since we build these networks from the ground up, the components are well-defined and amenable to mechanistic, predictive modeling. We are also using this combination of systems modeling and synthetic biology to build programmable cellular 'devices' that have applications ranging from drug delivery to tissue engineering.

Collaborators:
Jim Collins, Boston University; Rex Ahima, Penn; Vijay Kumar, Penn.