Fader Lab Post Doctoral Associate

The Faeder Lab seeks a postdoctoral associate to develop and apply mechanistic computational models of immune signaling and viral replication, and to advance the simulation methods that make such models practical. The initial assignment is to an NIH-funded collaboration with the laboratory of Dr. William Klimstra (Department of Immunology) that models Eastern equine encephalitis virus (EEEV) replication and its restriction by type I interferon. Depending on funding, the position will also support a collaboration with Dr. Rachel Gottschalk (Department of Immunology) on modeling cytokine-driven JAK-STAT signaling and its modulation by JAK inhibitors in human macrophages. A third area of work is the optimization of simulation capabilities for rule-based models, building on strong unpublished results in the lab. The postdoc will work closely with experimental collaborators and with a team of graduate students and will be co-mentored by Drs. Faeder and Gottschalk as the JAK-STAT project develops.

Viral replication and interferon restriction

• Refactor our published rule-based model of EEEV replication into a modular BioNetGen representation and recalibrate it using new high-resolution data (smFISH, ribosome profiling, ddRT-qPCR and reporter assays).

• Extend the model to include candidate interferon-mediated restriction mechanisms and use model-based analysis to identify the replication steps most sensitive to restriction.

• Perform Bayesian parameter estimation with regularization, including sensitivity, identifiability and uncertainty analyses, and generate testable predictions to guide experiments.

JAK-STAT signaling and JAK inhibitor response

• Develop, calibrate and analyze identifiable network models of cytokine (IL-6, IL-10 and type I interferon) signaling through JAK1, JAK2, TYK2, STAT1 and STAT3, including data-constrained representations of JAK contributions to STAT phosphorylation.

• Fit models to single-cell phospho-STAT measurements and extend them to population models of cell-to-cell and donor-to-donor variability using nonlinear mixed-effects methods, to predict individual sensitivity to selective JAK inhibition.

• Work with machine learning collaborators to integrate mechanistic models of STAT dynamics with sequence-to-function models of gene expression.

Simulation methods for rule-based models

• Develop, benchmark and optimize simulation algorithms and software for rule-based models, including network-based and network-free approaches, and contribute to the open-source BioNetGen software ecosystem.

• Prepare methods manuscripts and documentation describing new simulation capabilities and apply them to the lab’s modeling projects.

All projects

• Present results at lab and project meetings and scientific conferences, prepare manuscripts, and contribute to model and software documentation.

• Help mentor graduate and undergraduate students in the lab as appropriate.

Qualifications

• PhD (or near completion) in computational biology, systems biology, biophysics, physics, applied mathematics, computer science, statistics or a related quantitative field.

• Experience with ODE or stochastic modeling and parameter estimation; familiarity with Bayesian inference, sequential Monte Carlo or nonlinear mixed-effects methods is a strong plus.

• Programming proficiency in Python; experience with scientific computing, performance optimization or compiled languages (for example C++) is an advantage for the simulation methods work.

• Experience with rule-based modeling (BioNetGen, Kappa) or with modeling of intracellular signaling or viral replication is an advantage.

• Ability to work with experimental collaborators and to communicate quantitative results clearly to a mixed computational and experimental audience.

• Strong communication skills and a collaborative, critically engaged working style.

The University of Pittsburgh is an equal opportunity employer / disability / veteran.