Dr. Himel Mallick is an assistant professor of population health sciences in the Division of Biostatistics. Before joining Weill Cornell Medicine (WCM), he was an associate principal scientist at Merck Research Laboratories (MRL) and a postdoctoral fellow of computational biology and bioinformatics at the Broad Institute and Harvard University.
How did you first become involved in biostatistics?
I first became involved in biostatistics during my doctoral studies, where I worked on Bayes regularization in the context of high-dimensional variable selection for linear models. I developed a flexible framework by constructing a data-augmentation technique that facilitated efficient posterior computation, rich model summaries, and nuanced uncertainty quantification. Later, as a postdoctoral fellow, I developed mission-critical tools for the statistical analysis of microbiome multi-omics data. For example, I am the lead developer of popular bioBakery tools MaAsLin 2 and MelonnPan, which are regularly used in large-scale microbiome multi-omics studies.
Given my cross-disciplinary training in biostatistics and computational biology, my long-term career goal is to pursue translational research at the interface of multi-omics, machine-learning, and data science. I intend to continue my work on novel statistical methodology development and collaborative research in biomedical sciences.
Tell us about your research.
My research is largely focused on reverse translational efforts that aim to integrate vastly different kinds of biological data. I leverage machine learning, systems biology, and omics data science techniques to accelerate target identification and biomarker discovery across a range of indications.
To ensure relevance and impact, I create open-source software for analyzing high-dimensional data, such as next-generation sequencing and imaging data, and I promote the accessibility of my tools and methods through public repositories. I have served as a principal investigator in multiple industry projects and as a collaborator in large, multi-center studies, including the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study and the National Institutes of Health Integrative Human Microbiome Project (iHMP).
During my time at MRL, I spearheaded efforts in key microbiome, single-cell, spatial transcriptomics, imaging, and digital pathology projects. My work supported computational systems biology research within the company and expanded the omics data science portfolio with rigorous statistical and machine learning methodologies. This facilitated the translation of scientific discoveries into improved clinical and therapeutic outcomes.
What expertise do you bring to this role?
Over four rewarding years at Merck, I assumed lead statistician roles with numerous deliveries into therapeutic areas. My time there provided invaluable experience in large-scale data science management and integration. I understood the significance of statistical analysis plans in organizing computational biology projects and of artificial intelligence (AI), machine-learning, and statistics in delivering rapid results for biomarker and therapeutic intervention research. My work thus far has prepared me to guide students to explore careers in academia and industry. I am also eager to apply my experience in AI and machine-learning to real-world healthcare challenges at WCM.
What brings you to Weill Cornell Medicine?
Accepting a faculty position at WCM was a very easy decision. WCM offers an ideal academic and training environment for me to grow as a researcher by addressing critical biomedical science problems with innovative statistical methodologies. The resources available at NewYork-Presbyterian and the wider biomedical and life sciences communities in New York City provide a stimulating setting for methodological research. I am excited to be a part of quantitative, computational, and laboratory collaborations.
Are there any trends or issues you are currently following in your field?
Since ChatGPT's launch, we have entered a new era of AI-powered technologies that are revolutionizing various aspects of our lives. However, as AI technology progresses, concerns are raised about its long-term consequences, ethical implications, and privacy issues. For example, using ChatGPT and other AI tools can potentially exacerbate biases in data analysis and decision-making. I aim to identify areas where these rapidly evolving technologies can be most effectively utilized for the greater good, by examining their diverse applications and determining how ChatGPT can optimize efficiency and productivity while ensuring ethical and responsible use.