Chris Kieslich
Georgia Institute of Technology, GA·Computer Science
Vaccines and immune therapies often work by showing the immune system a short fragment of a pathogen's protein, called an epitope. Finding the right fragment is hard, because the body first has to chop the protein into pieces, then display a piece on a surface molecule, and finally have a T cell recognize it.
- Fields
Computer Science
Engineering
Microbiology & Immunology
- Topics
T-cell epitope prediction
antigen processing
MHC binding
- Grants
- 1 known grant, through Aug 2027
What they work on
- Build predictors of antigen processing that estimate which peptide fragments a cell actually produces when it degrades a protein, combining molecular models of the cutting machinery with statistical analysis of large public peptide datasets.
- Develop models of epitope recognition that predict binding of peptides to MHC molecules, the cell-surface proteins that display fragments, and to T-cell receptors, using molecular modeling to pinpoint the key contacts and machine learning to generalize.
- Quantify how mutations in pathogen and human sequences change antigen processing and epitope recognition, by mapping mutations onto the modeled contact positions and fitting the effects to publicly available immunological data.
- Create a probabilistic, scenario-based tool that samples antigen and host sequences from known mutation-prevalence distributions to estimate the likelihood that a candidate epitope will be generated and elicit a robust immune response.
More about the research
Each of those steps depends on genetics — both the pathogen's, which varies between strains, and the patient's, which varies between people. Prediction tools generally ignore that variability, and animal testing is limited because a mouse and a human cut proteins differently and recognize different fragments. This project builds computational tools that take both the pathogen sequence and the host's own genetic information as input.
Why it matters
Peptide-based vaccines can be produced quickly and stored for long periods, which matters when a new pathogen emerges, but they only work if the right pathogen fragments are chosen. Tools that account for genetic differences between pathogen strains and between individual people could make epitope selection more reliable, supporting faster vaccine design, personalized cancer vaccines, and identification of the specific fragments that drive allergic reactions.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Marmar Moussa
University of Oklahoma, OK·Computer Science·Early-career
Model and predict which T-cell receptors are activated by neoepitope vaccination in MethA and MC38 mouse tumor models, using single-cell TCR sequencing paired with DNA-barcoded peptide–MHC dextramers that report what each cell binds.
James Lee
New York University, NY·Microbiology & Immunology·Early-career
Map the complete route of peptide antigen loading onto MHC-I molecules using a chemical biology strategy applied both in living cells and in reconstituted systems rebuilt from purified components.
Hussein Sultan
Washington University in St Louis, MO·Microbiology & Immunology
Determine how the EBI3 subunit of the suppressive cytokine IL-35 drives induction of Tr1 regulatory CD4 T cells, using IL-35 reporter mice in established tumor models to identify which cells produce it.
The school
Georgia Institute of Technology: 50 labs on PI Pages, $1.5B a year on research, 572 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.