Special Issue: Statistical Methods, Algorithms and Applications in Biomedical Data Integration

Integrating multiple data sources has attracted wide interests as part of transition from data to knowledge with the potential to change the analytical scheme of modern biomedical research. In recent years, research initiatives have been created to leverage large-scale observational databases from multiple scientific disciplines and technologies, which present many significant methodological and computational challenges to traditional statistical methods and algorithms. Most notably, data collected from observational studies are leveraged for modern biomedical research to enrich study populations and improve controlling confounding factors. Despite infrastructure advancements, methodological and algorithmic challenges remain the barriers for data integration. Common analytical concerns in data integration include data storage and communication restrictions, statistical efficiency, protection of data privacy and against adversarial attacks, data harmonization over different formats, handling of missing data, heterogeneity across data sources, and external validity, among many others. Some new methods, algorithms and applications for data integration are being developed, but much remains unknown in terms of how well they perform or how they compare with conventional approaches. Moreover, with the ubiquitous availability of multi-source data and the increased desire to conduct research with massive data, new methodological and algorithmic developments are needed on many fronts of data integration, including uncertainty quantification, causal inference, and sparse and scalable analytical procedures.

Call for Papers Flyer: Statistical Methods, Algorithms and Applications in Biomedical Data Integration