AnniZ Lab

Our mission

• We aim to make microbes more useful to our society.
• We are decoding their behavior so we can design microbes to follow instructions:
• Sensing different environmental molecules and releasing different products accordingly.

What we do?

• We are building AI models starting from biological principles.
• We generate new training data by applying biological principles to data we already have.
• >0.5M microbial genomes and >6K microbial species
• 143K microbiome literature
• We train AI models on top to predict how proteins interact with DNA and ligands, to understand and control microbial behavior.
• More research and thoughts shared on LinkedIn and X!

Who are we?

• We are a team of microbiologists and software engineers previously from MIT and Google.

Our sevices

• What can we do?
• Technical R&D partner for companies in small molecule drug discovery, fermentation, probiotics, and agriculture.

• Core industrial domains
• Drug discovery: activating "silent" genes in microbes to discover new small molecule drugs
• Precision fermentation: designing microbes respond to environmental triggers to improve production yield
• Smart probiotics: instructing engineered probiotics where and when to release medicine
• Agriculture: engineering soil microbes for better plant growth and nitrogen fixation

Our tools

MicrobiomeConnectome: knowledge graph for druggable target prediction!

Biology-informed AI for DNA-protein binding predictions!

• New sequence aligner: Mapper
• Fast variant detector: QuickVariants
• Microbial AMR risk rank detector: arg_ranker
• Our github

AnniZ Lab Members

What to expect as our teammate

• Receive training in critical thinking and hypothesis testing
• Gain a solid foundation in statistics, microbiology, and AI computation
• Develop skills in coding, analysis, visualization, and public speaking
• Practice collaboration and communication
• More importantly, we care about your future! We are happy to mentor you in career development.

News!

• 2026 BIG NEWS: we are awarded SG$1M Singapore MOE Tier 2 grant for protein-DNA binding prediction!
• 2026 May Anni was invited to give a talk on Microbiome R&D and Business Collaboration Congress Asia.
• 2026 Apr Zhiqi talked about "Detection of HGT and selective sweeps in human gut microbiome via large data analysis and population genomics" in Bug seminars - great job.
• 2026 Mar Anni was invited as a working group member for National Infectious Disease Research Strategy, Singapore!
• 2026 Jan Dr. Xiaoqian Yu (University of Montreal) visited our lab at NTU.
• 2025 Dec Qiwen talked about "Biology-guided AI models to predict bacteria protein DNA binding" in BUG seminars - great job.
• 2025 Dec Dr. Klas Udekwu (University of Idaho & SLU) visited our lab at NTU.
• 2025 Nov Anni talked about "Uncovering diverse microbial mechanisms via computation" in SCELSE Retreat.
• 2025 Aug Jinxuan joined our lab as a PhD student. Welcome :)
• 2025 Aug Zhiqi switched to a PhD student. Congrats!
• 2025 July-Aug Aolin and Yifang transitioned to future paths. Good luck!
• 2025 July Anni gave a talk at Gordon Conference AEM 2025
• 2025 July Jeff presented work at Gordon Conference AEM 2025 :)
• 2025 July Anni presented work at Gordon Conference MPB 2025 :)
• 2025 June Anni gave a talk at MIT BE Seminar!
• 2025 March Aolin joined us as a research associate.
• 2025 Jan Qiwen joined us as a PhD student. Welcome to AnniZLab!
• 2025 Jan Zhiqi joined us as a research associate.
• 2024 Nov Yifang joined us as a research associate.

Highlighted discoveries

Case 1 of 6: published on Genome Biology, reported by NTU News Sequence aligners use fixed-size k-mers in their algorithms, with the choice of k-mer size significantly impacting alignment quality. Our tool, Mapper, overcomes this by dynamically incorporating k-mers of various sizes and gaps (gapped x-mers).

Case 2 of 6: published on Cell Genomics, reported by MIT News Gut microbiomes share CRISPR systems through horizontal gene transfer to update their phage defense.

Case 3 of 6: published on BMC Biology A novel algorithm for genetic variant identification that is 9x faster than existing tools, while producing fewer false positives and false negatives.

Case 4 of 6: published on Nature Communications Antibiotic resistance genes (ARGs) are everywhere, but most high-risk ARGs (Rank I, red bars) are only present in human-associated environments but not in natural environments.

Case 5 of 6: published on Microbiome Antibiotic resistance genes are highly prevalent (red and blue dots) in almost all bacterial genomes. Specifically, ARGs in Gammaproteobacteria (red dots) are spreading by horizontal gene transfer via class 1 integrons (i.e. intI1).

Case 6 of 6: Gut microbiomes are actively adapting in healthy human subjects through accumulating beneficial mutations in carbohydrate transport regulators.