AI-Powered Predictive Systems Biology

Focus on decoding the causal rules of life by interpretable AI models from multi-omic perturbation data to predict how complex tissues respond to disease and therapy.


We live in an era of exploding biological data.

We can now sequence genomes, profile single cells, and map entire tissues with extraordinary precision—giving us a detailed “parts list” of life itself.

But here’s the challenge: seeing the parts is not the same as understanding the system. Despite all this data, biology still lacks the ability to predict how complex systems behave. We know the components, but the blueprint remains hidden.

The next frontier in biology is clear: 👉 move from static maps to dynamic models that can learn, forecast, and explain life’s processes. This means shifting from correlation to causality, description to prediction.

Explore how Johnson’s research program is tackling this challenge— while also sharing insights into the cutting-edge biotechnologies shaping tomorrow’s breakthroughs.

Avatar

Qingzhou Zhang

A computational systems biologist

About

I am a computational systems biologist dedicated to building predictive models of life. I believe that the next era of biological discovery and therapeutic design will be driven by our ability to learn the causal rules that govern complex cellular systems.

My scientific blueprint is to build interpretable AI models from a synthesis of multi-modal data, from single-cell genomics to spatial proteomics. By focusing on perturbation-response experiments, my work moves beyond correlation to infer the causal logic of how cellular networks operate and adapt.

Interests

  • Predictive Systems Biology
  • Causal Network Inference
  • Cancer Therapeutic Resistance
  • Tissue Homeostasis & Repair
  • Interpretable Machine Learning

Education

  • PhD Biochemistry, 2016-2021

    University of Regina, Canada

  • MSc Bioinformatics, 2009-2010

    Univeristy of Manchester, England

  • BSc Biotechnology, 2004-2008

    Nanjing Agricultural University, P.R. China

Insights

Busting the Myth of the scRNA-seq "Dropout"

This analysis elegantly demonstrates that the high number of zeros in droplet-based scRNA-seq is not a technical artifact but is …

Run EPIC

The EPIC toolkit was initially published here: Hu, L. Z., et al. “EPIC: software toolkit for elution profile-based inference of …

Statistical Modelling of AP-MS Data (SMAD)

This R package implements statistical modelling of affinity purification–mass spectrometry (AP-MS) data to compute confidence scores to …