Disruptive AI Technologies for Molecular Biology and Medicine: DNA Motifs, CRISPR-Cas9 Off-Targets, and Cancer Screening from Blood

Speaker:  Ka-Chun Wong – Kowloon Tong, Hong Kong
Topic(s):  Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing

Abstract

In this talk, I will present my research group contributions in bioinformatics and health informatics in recent years. In particular, the unconventional and disruptive AI technologies are focused.

Firstly, the DNA binding of transcription factors is central to gene regulation and stem cell development. The DNA binding pattern (i.e. DNA motif) elucidation of transcription factors forms the basis for downstream research. Therefore, I will present our breakthroughs in elucidating DNA binding patterns from the protein-coding sequences of transcription factors using AI [1] as well as our synthetic biology approach to synthesize a heterodimeric DNA motif from two monomeric DNA motifs. A DNA motif published on Nature has been rescued [2].
Secondly, CRISPR-Cas9 is the predominant tool for gene editing and raised substantial concerns on its clinical implications. To avoid any side effect, its off-target predictions are fundamentally essential. I will present our recent work in predicting CRISPR-Cas9 off-targets using deep learning, the latest AI technology [3].
Finally, I will present our very recent work in screening cancers from blood. I will demonstrate how our proposed AI approach (CancerA1DE) can outperform the existing approach (CancerSEEK) proposed in John Hopkins University. In particular, our approach can double the existing sensitivity from 38% to 77% for the earliest cancer detection (i.e., Stage I) at the 99% specificity level [4].
[1] Wong, K. C. (2018). DNA Motif Recognition Modeling from Protein Sequences. iScience, 7, 198-211.

[2] Wong, K. C., Lin, J., Li, X., Lin, Q., Liang, C., & Song, Y. Q. (2018). Heterodimeric DNA motif synthesis and validations. Nucleic acids research, 47(4), 1628-1636.
[3] Lin, J., & Wong, K. C. (2018). Off-target predictions in CRISPR-Cas9 gene editing using deep learning. Bioinformatics, 34(17), i656-i663.
[4] Wong, K. C. et al. (2019). Early Cancer Detection from Multianalyte Blood Test Results. iScience.

About this Lecture

Number of Slides:  70
Duration:  45 minutes
Languages Available:  English
Last Updated: 

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.