Bo Wang

Bo Wang is a biomedical AI researcher whom Xaira Therapeutics promoted to Chief AI Scientist on July 6, 2026. His work connects single-cell modeling with efforts to predict how cells respond to genetic perturbations: he helped develop scGPT, which Latent Space describes as a starting point for Xaira’s X-Cell. Xaira credits him with pioneering X-Cell’s development and directing the integration of AI agents throughout its operations.[4][1]

Xaira announced Wang’s appointment as Senior Vice President and Head of Biomedical AI on April 3, 2025. His July 2026 promotion accompanied Ci Chu’s promotion to Chief Discovery Officer. The company jointly credited them with contributions to X-Cell and the X-Atlas/Pisces dataset that powers it, identifying Chu’s leadership of high-throughput biology alongside Wang’s model-development work.[3][4]

X-Cell is a diffusion language model designed to predict genome-scale changes in gene expression following genetic perturbations. Its documentation describes training on 25.6 million perturbed single cells across seven CRISPRi Perturb-seq screens and incorporating multimodal biological priors through cross-attention. This gives the project a different task from simply describing observed cell states: predicting what changes after an intervention.[7][1]

The connection to scGPT is concrete at the model-variant level: X-Cell’s documentation lists the 55-million-parameter X-Cell Mini as initialized from scGPT. Latent Space’s interview with Wang and Chu, published in July 2026, places this work within their teams’ effort to generate information-rich experimental data for perturbation prediction.[7][1]

Wang expressed his standard for that work in Xaira’s promotion announcement: “AI only matters here if it makes science better.” He presented X-Cell as evidence that the company could meet that standard.[4]

Before his Xaira leadership roles, Wang worked on computational biology, including cancer subtype prediction and single-cell analysis during his Stanford doctorate. His earlier publications provide context for his later biological AI work through research on combining genomic data and analyzing single-cell measurements.[5][8]

Wang coauthored the 2014 Nature Methods paper “Similarity network fusion for aggregating data types on a genomic scale” with Aziz M. Mezlini, Feyyaz Demir, Marc Fiume, Zhuowen Tu, Michael Brudno, Benjamin Haibe-Kains and Anna Goldenberg. His publication record also lists a 2017 Nature Methods paper on visualization and analysis of single-cell RNA-seq data using kernel-based similarity learning, coauthored with Junjie Zhu, Emma Pierson, Daniele Ramazzotti and Serafim Batzoglou.[8]

Wang earned his PhD in Computer Science at Stanford University in 2017. At the University of Toronto, he belonged to Medical Biophysics and was cross-appointed in Surgery before joining Laboratory Medicine & Pathobiology. Toronto’s faculty profile lists joint appointments in Laboratory Medicine & Pathobiology and Computer Science, a Chief AI Scientist role at University Health Network, and a Canada CIFAR AI Chair at the Vector Institute.[5]

His Toronto work also includes medical AI education. The university credits Wang with expanding AI education in Laboratory Medicine & Pathobiology alongside Rahul G. Krishnan and launching a graduate course, Basic Principles of Machine Learning in Biomedical Research.[5]