The Department of Biostatistics and Health Data Science is thrilled to announce that PhD student Haoran Hu has made a significant contribution to the field of single-cell research with the publication of a paper titled "A unified model-based framework for doublet or multiplet detection in single-cell multiomics data" in the top research journal Nature Communications.
Haoran proposes a compound Poisson framework to model the formation process of multiplets and derives statistical inference to predict the multiplet status of each droplet. This research represents a significant advancement in the field, offering a comprehensive approach to detecting doublets or multiplets in complex single-cell multiomics datasets. The paper has also been chosen as a featured article by the editor and highlighted in “Biotechnology and methods” section https://www.nature.com/collections/idhhgedgig.
Haoran explains, “In single-cell research, the inevitable issue of multiplets, where two or more cells are encapsulated within a single droplet, can lead to spurious cell type annotations and obscure true biological findings. This work introduces COMPOSITE, a data-driven approach to address the fundamental problem of multiplets in single-cell research, especially in single-cell multiomics settings. In this work, we proposed a compound Poisson framework to explicitly model the formation process of multiplets and derived statistical inference to predict the multiplet status of each droplet.” Haoran's paper has the potential to significantly improve the accuracy and reliability of single-cell research, enabling scientists to gain a deeper understanding of biological systems and driving new discoveries in fields such as medicine and genomics.
“This work demonstrates the cutting-edge opportunities available in our department. The strong mentorship and collaborative environment, which are hallmarks of our PhD program, made this achievement possible. I extend my gratitude to Professor Ying Ding for her expert guidance and mentorship throughout the project, as well as to former PhD student Xinjun Wang and current PhD students Molin Yue and Lang Zeng for their critical insights and assistance in data analysis. I would also like to acknowledge physician scientist Dr. Richard H. Duerr from UPMC, whose collaboration was instrumental in generating advanced single-cell multiomics data and conducting validation experiments. Special thanks go to Professor Wei Chen, my PhD advisor, and the corresponding author of the article, for his invaluable support and mentorship." Haoran invites prospective students to explore the exciting research happening in our school, where collaboration and innovation drive meaningful discoveries.
Please join us in congratulating Haoran on this outstanding achievement!
-Calvin Dziewulski