This talk will start discussing a typical computational workflow to identify suitable vaccine antigens for a disease X. We also propose a new data integration approach for monitoring influenza antigenic change efficiently. To achieve this, we stratify influenza surveillance sequences into closely-connected transmission clusters via genetic distance and analyze clusters for enrichment of epidemiological and geo-temporal data accessible through an interactive webtool for near-real-time surveillance/analysis also by non-computer-experts and without exposing potentially sensitive clinical data. FluCluster-AI enables real-time detection of emerging variants both in local and global context through connection with the GISAID platform and provides a privacy-preserving option to find correlations of any user-provided meta data labels with variants and specific mutations. We provide examples for automated analysis of HI titre data to pinpoint mutations driving influenza antigenic drift.