SAN FRANCISCO, California, USA, 8 July 2025 – In a comprehensive Genomic Press Interview published in Brain Medicine, Dr. Michael C. Oldham shares his unconventional journey from advertising executive to computational neuroscientist and his groundbreaking contributions to understanding the human brain's cellular and molecular architecture through gene coexpression analysis.
From Madison Avenue to molecular neuroscience
Dr. Oldham's path to neuroscience was anything but direct. After graduating from Duke University at age 20 with a pre-med focus, he found himself unable to commit to medical school, recognizing he lacked the intrinsic desire to treat patients. Following a stint in San Francisco's advertising industry during the dot-com boom, his fascination with human language evolution and what distinguishes human brains from those of our closest primate relatives led him back to academia.
"The genetic changes that gave rise to the modern human brain were the catalyst for life as we know it," Dr. Oldham reflects in the interview. This fundamental question drove him to pursue a PhD at UCLA, where he would make discoveries that continue to shape neuroscience research today.
Pioneering gene coexpression network analysis
Working with Dr. Dan Geschwind at UCLA and biostatistician Dr. Steve Horvath, Dr. Oldham performed the first genome-wide analysis of transcriptional covariation in the human brain. His eureka moment came when he realized that recurrent patterns of gene activity in brain samples corresponded to transcriptional signatures of different cell types.
"Variation in the cellular composition of bulk tissue samples should inevitably drive the covariation of markers for different cell types," Dr. Oldham explains. This insight, published in Nature Neuroscience in 2008, demonstrated how gene coexpression analysis could reveal optimal markers of cell types and states—a principle that still forms the central thesis for his laboratory at UCSF.
The approach, known as Weighted Gene Coexpression Network Analysis (WGCNA), has become a cornerstone technique in genomics research. Unlike traditional differential expression analysis that compares individual genes between cohorts, WGCNA identifies robust patterns of coordinated gene activity within biological systems. This methodology has proven particularly powerful for understanding complex tissues like the brain, where multiple cell types interact in intricate ways.
From brain evolution to brain tumors
Dr. Oldham's early research focused on analyzing patterns of gene activity in the brains of humans and other species. These efforts identified functionally significant gene expression changes in human radial glia (Nature, 2014), interneurons (Cerebral Cortex, 2018), and astrocytes (Nature Neuroscience, 2018), while introducing novel methods for aggregating and comparing patterns of gene activity among biological systems.
More recently, his research focus has shifted from studying what makes human brains unique to tackling one of medicine's most challenging diseases: malignant gliomas. As a faculty member in UCSF's Department of Neurological Surgery and Brain Tumor Center, he applies his computational approaches to these notoriously heterogeneous brain tumors.
His team has analyzed gene activity patterns from over 17,000 human brain samples, including approximately 10,000 normal and 7,000 malignant glioma samples. This massive undertaking has led to the development of OMICON (theomicon.ucsf.edu), a platform designed to make the patterns of gene activity in these complex datasets accessible to the broader research community. The resource contains over 100,000 gene coexpression modules that have been extensively characterized via enrichment analysis with thousands of curated gene sets, providing researchers worldwide with unprecedented insights into brain function and dysfunction.
By comparing patterns of gene activity between normal human brains and malignant gliomas, Dr. Oldham and his team are pinpointing highly reproducible molecular changes in specific cell types of the glioma microenvironment, including vascular cells and neurons. These molecular signatures provide opportunities for developing novel biomarkers and targeted treatment strategies for glioma patients. For example, cell-surface markers of glioma vasculature provide a potential molecular 'zip code' for targeting gliomas via the bloodstream.
Confronting the reproducibility crisis
Beyond his primary research, Dr. Oldham has become increasingly concerned with what he describes as science's reproducibility crisis. "If most of the findings we toil to produce cannot feasibly be reproduced, what is the point?" he asks, highlighting a challenge that extends far beyond neuroscience.
His response has been to take leadership roles addressing these systemic issues. As Vice Chair of UCSF's Academic Senate Committee on Library and Scholarly Communication, he has launched a pan-UCSF Task Force on research data and metadata standardization. While the topic might sound technical, Dr. Oldham emphasizes its critical importance: these standards are essential prerequisites for more open and reproducible science, more precise biomedical knowledge representation, and more efficient collaboration.
"Although there are many factors that affect the reproducibility of published research findings, there is no reason in principle why data analysis should not be completely reproducible," Dr. Oldham notes. "By standardizing how we package and describe our research data, we can accelerate data discovery and analysis, including the use of artificial intelligence. More generally, standardized data packages with persistent identifiers can serve as building blocks for new technology infrastructure to modernize scholarly communication around reproducible data analysis."
The human side of scientific discovery
The interview reveals personal insights that shaped Dr. Oldham's career trajectory. His decision to spend two additional years in graduate school after his first major publication—a choice some considered "nuts"—resulted in a second, even more impactful paper that secured his selection as a UCSF Sandler Faculty Fellow. This prestigious position provided him with immediate independence and funding to establish his own laboratory.
When not advancing neuroscience, Dr. Oldham can be found on the trails of Marin County, where he lives, often walking alone and lost in thought. He maintains close friendships from his San Francisco advertising days, adhering to their motto: "ABC (always be celebrating!)."
Looking ahead, Dr. Oldham sees the integration of multiscale and multimodal data as crucial for understanding brain complexity. He advocates for standardized data production strategies that leverage robotic automation to generate reproducible datasets at scale. Dr. Oldham also believes that neuroscientists must 'flip the switch' from descriptive analysis of biological systems to predictive analysis using statistical models. "There is a big difference between describing what you think a dataset means versus predicting what you will see in the next dataset," he says.
Dr. Michael C. Oldham's Genomic Press interview is part of a larger series called Innovators & Ideas that highlights the people behind today's most influential scientific breakthroughs. Each interview in the series offers a blend of cutting-edge research and personal reflections, providing readers with a comprehensive view of the scientists shaping the future. By combining a focus on professional achievements with personal insights, this interview style invites a richer narrative that both engages and educates readers. This format provides an ideal starting point for profiles that explore the scientist's impact on the field, while also touching on broader human themes. More information on the research leaders and rising stars featured in our Innovators & Ideas – Genomic Press Interview series can be found in our publications website: https://genomicpress.kglmeridian.com/.
The Genomic Press Interview in Brain Medicine titled "Michael C. Oldham: Clarifying the cellular and molecular architecture of the human brain in health and disease through gene coexpression analysis," is freely available via Open Access on 8 July 2025 in Brain Medicine at the following hyperlink: https://doi.org/10.61373/bm025k.0080.
About Brain Medicine: Brain Medicine (ISSN: 2997-2639, online and 2997-2647, print) is a peer-reviewed medical research journal published by Genomic Press, New York. Brain Medicine is a new home for the cross-disciplinary pathway from innovation in fundamental neuroscience to translational initiatives in brain medicine. The journal's scope includes the underlying science, causes, outcomes, treatments, and societal impact of brain disorders, across all clinical disciplines and their interface.
Visit the Genomic Press Virtual Library: https://issues.genomicpress.com/bookcase/gtvov/
Our full website is at: https://genomicpress.kglmeridian.com/
