Park Lab

You will know the truth, and the truth shall set you free. (John 8:32)
PARK LAB
Updated HuGen Logo

Our Vision: A.I. Meets Biology

We develop data-science techniques, AI-driven tools, and statistical inference methods and run them on public data to elucidate large-scale biological dynamics for complex diseases, such as cancer and sepsis. Based on the dynamics elucidated, we collaborate with clinical scientists/experimental biologists to develop 'precision medicine' strategies for the patients. Through the steps, we envision revealing mechanistic and translatable insights into improving clinical outcomes of complex diseases by decoding hidden patterns, extracting relationships between complex attributes, identifying succinct characteristics hidden in large volumes of biomedical data across multiple regulatory layers.

Our Research

Decoding molecular interactions in complex diseases

Decoding molecular interactions in complex diseases using network modeling

Relatively little is known about the RNA-level (post-transcriptional) regulation in diverse diseases partly due to the lack of computational models that can decode the complex interactions. We develop network modeling combined with statistical learning models to elucidate the novel roles of the complex interactions.

Innovating clinical care for sepsis using deep learning

Innovating clinical care for sepsis using deep learning

Sepsis, defined by suspected infection and development of organ failure contributes to 1 of 5 deaths globally, with the majority of these deaths observed in infants and children. In the collaboration network from 24 hospitals across the nation, we develop 'precision medicine' strategies to treat sepsis by developing machine-learning and causal inference methods.



PROJECT WEBSITE
Tracing viral infection using single-cell RNA-Seq data

Advancing cancer immunotherapy using single-cell multi-omics analysis

Although cancer immunotherapies provide long-term clinical benefits in general, many patients still do not get benefits from the current therapy regimens. In collaboration with the melanoma program leader at the Hillman Cancer Center, we develop bioinformatic methods that analyze single-cell multi-omics data to improve the efficacy of the therapies.