16 March 2010
Genedata has released Genedata Analyst, an integrated statistical and data analysis platform with advanced visualisation capabilities. The software is built on a client-server architecture and can handle huge and complex data sets of billion-plus data points while supporting small research groups or hundreds of users. Simultaneously managing data from different sources, it gives researchers and biostatisticians a secure and scalable data-mining platform, which can be integrated tightly with existing research IT ecosystems.
Genedata Analyst gives a broad range of scientists the capabilities for sophisticated data analysis. Combining the management of diverse data sources and types with data mining tools and knowledge management systems, the platform provides a comprehensive set of statistical approaches and interfaces to third-party algorithms. Features include context-sensitive help, guided workflows and carefully chosen default settings to provide enhanced usability, enabling non-expert users to access, use and benefit from statistical tools for data integration, mining and interpretation.
The platform is scalable and can process large and complex datasets of many measurements, design factors, and covariates. More than a billion data points are supported with no limit on the number of observables, experiments or covariates. It provides advanced visualisation and standard tools include histograms, bar charts, box plots, heat maps, scatter plots, trees, maps and parallel coordinate plots. Established general-purpose tools are provided, such as ANOVA, linear models, principal components and partial least square analysis, self-organising maps (SOM) and clustering methods together with advanced machine learning algorithms (LDA, SVM, KNN) for classification and prediction and tools for data normalisation, transformation and imputation.
The platform is designed so that a client-server application can support up to hundreds of users on standard hardware. Open technologies provide low TCO through public APIs that enable integration with in-house data warehouses, public data repositories, and third-party applications.