February 26, 2026 Webinar
Evidence-Based Practice for microRNA-seq Data Harmonization: Contextual Evaluation and Robust Benchmarking
Li-Xuan Qin, PhD
Abstract:
Reliable translation of omics data depends on effective harmonization that mitigates artifacts from variable experimental handling. Although many methods exist (spanning normalization and batch-effect correction), their downstream consequences have mostly been assessed for differential expression and often on oversimplified benchmarks. I will present a contextual, systematic framework for evaluating harmonization that integrates AI-augmented benchmark datasets, reproducible computational pipelines, and accessible software tools. Using microRNA-seq as a case study, we assess harmonization performance in two analysis contexts – sample clustering and sample classification – each paired with multiple analytical algorithms. The results clarify when simple scaling methods suffice, when more elaborate methods help (or hurt), and how method choice should align with the downstream task and algorithm. We corroborate our findings with publicly available real-world datasets and share open software to reproduce and extend our research. Collectively, this study underscores the need for context-specific harmonization and offers concrete guidance on effective harmonization-analysis pairings to advance translational genomics.
Short Bio:
Dr. Li-Xuan Qin is an attending biostatistician at Memorial Sloan Kettering Cancer Center, New York, NY. She develops benchmark data, analytic methods, and computational tools for enabling reproducible statistical translations of cancer omics data. Her recent work studied the issue of data harmonization for microRNA and RNA sequencing, which illustrated its dependence on the analysis goal (such as sample classification and survival prediction) and called for evidence-based practice. Her research program has been supported by multiple grants awarded by the NIH. At MSK Dr. Qin collaborates with multidisciplinary investigators engaged in research on soft tissue sarcoma and colorectal cancer, for design and analysis of retrospective and prospective studies. She serves as Co-Director of the Biostatistics and Bioinformatics Core of the MSK Soft Tissue Sarcoma SPORE (Specialized Program of Research Excellence).
Dr. Qin has published more than 100 peer-reviewed research articles in prestigious statistical, bioinformatics, and medical journals, including Biometrics, Nature Methods, Nucleic Acids Research, Briefings in Bioinformatics, Nature Genetics, Lancet Oncology, JAMA Oncology, and Journal of Clinical Oncology. Additionally, she serves as Handling Editor for Briefings in Bioinformatics, Associate Editor for Statistical Applications in Genetics and Molecular Biology, and Council Representative for the Section on Statistics in Genomics and Genetics of the American Statistical Association. She has also served as a standing member of the NIH ASPA and BMRD study sections, as well as a grant reviewer for the NSF, the French National Research Agency, and Cancer Research UK. Dr. Qin has organized invited and topic-contributed sessions for JSM and ICSA for over 10 years.
Hope to see you at the webinar!