Computational Insights into Human Evolution from FAU PhD Graduate Sandipan Paul Arnab

Sandipan Paul Arnab recently earned his Ph.D. in Computer Engineering from the Department of Electrical Engineering and Computer Science at FAU’s College of Engineering and Computer Science. A former member of Professor Michael DeGiorgio’s lab, he represents the next generation of researchers bridging computational innovation and biological discovery. His academic journey reflects a deep integration of engineering principles, machine learning, and evolutionary biology — a path that now leads him to a postdoctoral position at UC Berkeley.

His interest in computational genomics was sparked during his early career as an IoT engineer in Bangladesh, where he worked on a machine learning project that led him to pursue data-driven research. After completing his bachelor’s degree in Electrical and Electronic Engineering from BRAC University, he joined FAU for graduate studies, where he pursued both a master’s in computer science and a PhD in computer engineering. At FAU, his work focused on developing novel computational tools to detect natural selection in genomic data, an inherently complex challenge due to sequencing noise, population structure, and the overlapping signatures of different evolutionary processes.

Throughout his doctoral research, Sandipan explored whether rich genomic signals could be distilled into compact representations suitable for machine learning and signal processing. His work introduced scalable, interpretable frameworks that leverage advanced methodologies, including convolutional neural networks, support vector machines, fast Fourier transforms, and wavelet decomposition, to extract meaningful evolutionary insights. He also applied image processing techniques to transform genomic data into structured formats for classification.

Sandipan’s efforts have resulted in five publications in top-tier journals, including Molecular Biology and Evolution, Philosophical Transactions of the Royal Society B, and Evolution Letters. These papers highlight advances in detecting selective sweeps across diverse human populations, including groups that have been historically understudied in genomic research. His research has already begun to generate impact: in ongoing work, he is applying these tools to genomic data from the Bengali population in Bangladesh, uncovering signals of ancestral introgression that deepen our understanding of regional evolutionary history.

The research was conducted under the supervision of Dr. DeGiorgio, in collaboration with an international, interdisciplinary team, including Dr. Matteo Fumagalli from Queen Mary University of London and several postdoctoral and graduate researchers at FAU. Sandipan also made a point to ensure his work remains transparent and accessible; all code developed during his projects is openly available on GitHub. He has presented his work at leading conferences and institutions, including the University of Pennsylvania, UC Berkeley, and the American Society of Human Genetics.

Sandipan’s research directly aligns with the Center for SMART Health's mission by demonstrating how data science and intelligent modeling can enhance our understanding of human health. By translating genomic complexity into actionable insights, his work paves the way for future applications in medical genomics, particularly in understanding population-specific disease susceptibilities.

Now moving on to postdoctoral research at UC Berkeley, Sandipan will continue to build on this foundation, developing methods to analyze both present-day and ancient DNA in order to illuminate the deep history of human adaptation and its biomedical relevance. His journey exemplifies the kind of interdisciplinary and impactful research the Center for SMART Health seeks to foster.