KUEI-YUEH (Kuei) KO, PhD Candidate
Computational Biologist & Biostatistician

DUKE UNIVERSITY, Durham, NC

​

PhD Project: Functional characterization of regulatory regions using genome-wide functional screens and 3D chromatin interactions.

• Systematically compared high-throughput reporter assays and CRISPR screens to study the concordant and discordant signals in measured regulatory effects across different functional characterization assays.

• Integrated a wide range of functional annotations for genome-wide analysis, incorporating ChIP-seq results and candidate cis-regulatory elements from the ENCODE database.

• Actively collaborated with multiple labs as part of an ENCODE working group, playing a pivotal role in designing rigorous statistical tests and bridging the gap between functional characterization assays and chromatin organization data.

​

PhD Project: Combinatorial analysis of regulatory elements using STARR-seq data.

• Modeled regulatory activities in STARR-seq data to investigate the cooperative and combinatorial effects of transcription factor binding to DNA.

• Employed generalized linear models to estimate the regulatory impacts of motifs and their interactions within STARR-seq libraries, identifying transcription factor motifs with significant effects, consistent with previous studies.

• Leveraged deep sequenced libraries from STARR-seq data, pioneering a neural network approach to understand the impact of varied motif arrangements on regulatory activities.

​

Master Project: Classification and feature extraction of single-cell data using deep learning-based approach.

• Aimed to automate the identification of rare cell populations in single-cell data and extract predictive, biologically relevant features for these cellular subsets.

• Utilized deep learning image classification models in order to analyze cell distribution in flow cytometry data and to uncover potential gene interactions within identified cell subpopulations.

​

BCTIP Research Intern.

• Collaborated with statistical researchers to help evaluate their cluster registering method.

• Developed an interactive user interface using R Shiny and Flexdashboard to demonstrate their methods.

​

Teaching Assistant, High-throughput Sequencing Summer Course.

• Helped implement coding tutorials and delivered a lecture on pathway analysis to individuals with a biology background.

• Assisted students in data analysis by clarifying lecture content and converting their ideas into code, receiving consistent positive feedback.

• Acquired proficiency in Docker and Sphinx for setting up development environments, ensuring research reproducibility, and creating code documentation (https://duke-hts-2019.netlify.app/index.html).

​

​