PeopleofSuperintelligence

Researcher

Chi-Man Liu

Also known as Chi Man Liu

Chi-Man Liu is a computer scientist whose work applies parallel computing and AI to genomics. He earned a BSc in computer science and a PhD from HKU and an MS from the University of Wisconsin–Madison, and coauthored GPU tools for read alignment and genome assembly. In 2020, he coauthored deep neural network research on germline variant calling; his later work includes a Chinese pan-genome study and AutoPM3, an LLM method for extracting variant evidence from scientific papers. He returned to HKU as a Lecturer in 2025 after more than ten years in bioinformatics, and is a Lecturer in the Division of AI & Data Science, where he taught an AI literacy course to alumni in February 2026.

Timeline

  1. 2012

    SOAP3 for short reads

    In 2012, he coauthors SOAP3, a GPU-based parallel aligner for short DNA reads.

  2. 2014

    CX1 and MEGAHIT

    In 2014, he coauthors CX1, which uses GPU, CPU, and cluster parallelism to construct Burrows–Wheeler indexes, and MEGAHIT, a single-node assembler for large metagenomic datasets.

  3. 2020

    Deep neural variant calling

    In 2020, he coauthors a Nature Machine Intelligence study of deep neural networks for germline variant calling from pileup data. The study presents Clair, a caller for single-molecule sequencing data.

  4. 2021

    A Chinese pan-genome

    In 2021, he coauthors a Chinese pan-genome study using 486 Han Chinese genomes and identifying 276 Mbp of sequences missing from the human reference. Adding common sequences improves variant calling and mapping.

  5. 2024

    AutoPM3 evidence extraction

    In 2024, he coauthors the AutoPM3 preprint, which uses open-source large language models and retrieval-augmented generation to extract PM3 evidence from scientific literature.

  6. By 2025

    HKU and Wisconsin degrees

    By 2025, he holds a BSc in Computer Science and a PhD from HKU, plus an MS from the University of Wisconsin–Madison.

  7. 2025

    Returning to HKU

    In 2025, he returns to HKU as a Lecturer in AI & Data Science after more than ten years in the bioinformatics industry, working as a researcher, software engineer, and project manager.

  8. As of October 2026

    AI literacy at HKU

    He is a Lecturer in HKU's Division of AI & Data Science. In February 2026, he teaches a four-week AI literacy mini-course for alumni, introducing modern AI and practical tools for everyday work.

Career

  1. NowUniversity of Hong KongLecturer

Education

  1. University of Hong KongBSc in Computer Science
  2. University of Wisconsin–MadisonMS

From the biography.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes University of Hong Kong. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes University of Hong Kong. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes University of Hong Kong. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  4. 5 of the 25 people with the most-cited recent papers also lead on career citationsIncludes University of Hong Kong. We compare citations to 2023–2025 papers with career citations and h-index, after correcting reviewed publication-date errors: who leads, when they started and where they work.

Sources and credits

Photo: We have no published photo of Chi-Man Liu yet. Send us a free-licensed photo.

  1. Liu, Chi Man ai.hku.hk. Checked 20 October 2026.
  2. GPU-Accelerated BWT Construction for Large Collection of Short Reads arxiv.org. Checked 20 October 2026.
  3. MEGAHIT: An ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph arxiv.org. Checked 20 October 2026.
  4. Exploring the limit of using a deep neural network on pileup data for germline variant calling nature.com. Checked 20 October 2026.
  5. Building a Chinese pan-genome of 486 individuals nature.com. Checked 20 October 2026.
  6. AutoPM3: enhancing variant interpretation via LLM-driven PM3 evidence extraction from scientific literature sciety.org. Checked 20 October 2026.
  7. CDS Welcomes CS Alumni Reunion and Launches AI Literacy Mini-Course cds.hku.hk. Checked 20 October 2026.