Flow cytometry data is increasingly complex, with more than 30 parameters measured in some studies. Large databases, such as ImmPort, provide a rich resource of potential biological information. However, extracting the full biological information from these data is very time-consuming by traditional manual methods, which do not fully explore this multi-dimensional data. Several algorithmic approaches have been developed to facilitate this analysis, and although some of these have been made available in a user-friendly format, many of the methods are still difficult to access without considerable computational expertise. This application focuses on a suite of several tools for pre-processing, clustering, analysis and exploration of high-dimensional flow cytometry data. Pre-processing tools include a registration method that can selectively remove one source of variation, such as batch effects between different experiments, while leaving other sources of variation, such as experimental group differences, intact for further analysis. The SWIFT clustering algorithm was developed as a high-resolution analysis method that identifies a large number of potential sub-populations. SWIFT is also very sensitive, and can detect rare sub-populations at less than one part per million. SWIFT cluster templates assign events in many samples to the same clusters, thus facilitating comparisons between experimental groups and manipulations. As the resulting processed data is still highly complex, several further tools have been developed to facilitate flexible and rapid exploration of the cluster data. These include tools for rapid detection of anomalies in the data, and rigorous comparison between samples. Other tools visualize complex properties of clusters in simple, interactive graphics that encourage exploration of novel cell populations. ImmPort Galaxy is a web-based platform designed to host multiple analysis tools that can be applied to data in ImmPort and other databases. This application builds on the establishment of the ImmPort Galaxy web- based platform, to further develop the SWIFT suite of programs and integrate these into ImmPortGalaxy. Several independent tools will be integrated, with standardized data types that allow different pre-processing, population identification, cluster and sample classification, and analysis/exploration tools to be used interchangeably. This will allow each of the existing ImmPort Galaxy tools, and the new tools, to all be used at maximum effectiveness by choosing the best tool for the specific dataset and question, for each step in the data processing pipeline.
Flow cytometry produces large amounts of highly complex data on cell populations, and the accumulation of this data has so far outstripped the capability to extract the full biological meaning from the data. Research computational tools have been constructed for flexible and rapid analysis and exploration of flow cytometry data, and it is now proposed that these tools should be transformed into user-friendly, interactive web-based tools to broadly facilitate the exploration of these rich datasets.
