Advanced 3GS-based bioinformatics algorithms and a complete bioinformatics solution for clinical genetics
Grant Data
Project Title
Advanced 3GS-based bioinformatics algorithms and a complete bioinformatics solution for clinical genetics
Principal Investigator
Professor Lam, Tak Wah
(Project Coordinator (PC))
Co-Investigator(s)
Dr Luo Ruibang
(Co-Investigator)
Dr Lo Fai-Man Ivan
(Co-Investigator)
Dr Chan Tsun Leung
(Co-Investigator)
Dr Luk Ho-Ming
(Co-Investigator)
Professor Cheung David Wai Lok
(Co-Investigator)
Duration
18
Start Date
2018-11-01
Completion Date
2020-04-30
Amount
603000
Conference Title
Advanced 3GS-based bioinformatics algorithms and a complete bioinformatics solution for clinical genetics
Keywords
Advanced 3GS-based, bioinformatics algorithms, bioinformatics solution, clinical genetics
Discipline
Others - Computing Science and Information Technology
Panel
Engineering (E)
HKU Project Code
PiH/160/18
Grant Type
Postdoctoral Hub Programme for ITF projects
Funding Year
2017
Status
Completed
Objectives
OBJECTIVES: Building from our success in NGS bioinformatics & the new-to-market 3GS-sequencer MinION, we team up with medical doctors to develop the first bioinformatics solution for analyzing 3GS data of DNA to give a complete & precise clinical diagnosis of genetic diseases. Although existing NGS analysis performs well on diagnosing point mutations (single-nucleotide variations, SNV), it often fails to detect structural variations (SV) that cause common genetic diseases (e.g., Huntington's disease, spinocerebellar ataxia caused by trinucleotide-repeat in GC-biased regions; retinitis pigmentosa, neurofibromatosis-1 caused by long insertions). Thanks to MinION's advantages of negligible fixed cost, longer read length and no GC- bias, our 3GS bioinformatics solution delivers affordable and complete diagnosis. It not only supports precise diagnosis of SV for common genetic diseases, but also matches NGS bioinformatics' performance on diagnosing SNV. R&D: 3GS-data analysis is indeed challenging as they are highly-erroneous than NGS data (10+% vs 0.5%). We will revolutionize DNA-alignment algorithm to truly reflect 3GS's error model, develop machine-learning models for specific SV-SNV detection, extend clinical interpretation system, and design dedicated wet-lab supports for 3GS DNA capturing.