Mehran Karimzadeh

I am currently an AI Scientist at Xaira Therapeutics and am based in Seattle, WA. I am interested in leveraging deep learning for personalized medicine applications, particularly in the context of early detection, target discovery, and precision therapy. I have contributed to deep learning models for liquid biopsy and cellular perturbation, including Orion, Exai-1 (cover article, Nature Machine Intelligence 2025), and X-Cell (pre-print 2026).

Curriculum Vitae →

Select Publications

A curated list of manuscripts where Mehran is the first or co-first author, showcasing key research contributions.

Publication representation
X-Cell: Scaling Causal Perturbation Prediction Across Diverse Cellular Contexts via Diffusion Language Models.

bioRxiv. • 2026

Publication representation
A multimodal cell-free RNA language model for liquid biopsy applications.

Nature Machine Intelligence • 2025 • Cover Article

Publication representation
Detection of early stage colorectal cancer using cell-free oncRNA biomarkers and AI.

Clinical Cancer Research. • 2025

Publication representation
Deep generative AI models analyzing circulating orphan non-coding RNAs enable accurate detection of early-stage non-small cell lung cancer.

Nature Communications • 2024

Publication representation
Multi-modal single-cell atlas of the human breast upon androgen therapy

Cell Genomics • 2023

Publication representation
Human papillomavirus integration transforms chromatin to drive oncogenesis

Genome Biology • 2023

Publication representation
Virtual ChIP-seq: predicting transcription factor binding by learning from the transcriptome.

Genome Biology • 2022

Publication representation
Umap and Bismap: quantifying genome and methylome mappability.

Nucleic Acids Research • 2018

Publication representation
Aberration hubs in protein interaction networks highlight actionable targets in cancer.

Oncotarget • 2018

Publication representation
Top considerations for creating bioinformatics software documentation.

Briefings in Bioinformatics • 2018

Class Presentations

DNA methylation workshop

Genomic Methodologies, University of Toronto
Isoform analysis workshop

Genomic Methodologies course, University of Toronto
Introduction to biostatistics

Department of Medical Biophysics, University of Toronto (2018)
Introduction to biostatistics

Department of Medical Biophysics, University of Toronto (2019)
Introduction to ggplot

MBP Tech talks, University of Toronto
Introduction to R

MBP Tech Talks, University of Toronto

Blog

Thoughts on AI, bioinformatics, and scientific computing.

Interpretability of Deep Learning Models: A Historical Review
Apr 30, 2026

From Gradient Saliency to Sparse Autoencoders (2013–2026). A comprehensive review of the most important methods in Explainable AI.

By Mehran Karimzadeh

View all posts →

Contact

mehran dot karimzade at gmail dot com