The design and development of the XB Bio-Integration Suite is based upon a visionary concept. We believe that in the future, patients will be assessed for their risk of developing a disease, accurately diagnosed at an early point after disease initiation, and then treated with "personalized" therapies based upon the specific genomic and phenotypic makeup of both the patient and his diseased tissue(s).
To reach this goal, XB Bio-Integration Suite was designed as a fully integrated biological and medical informatics solution that allows users to generate, refine, and validate causal hypotheses across disparate data sets.
The core of the XB Bio-Integration Suite is a platform built on a set of open standards-based data access tools with associated analysis and annotation software.
Data Analysis: The XB Bio-Integration Suite analyzes data from a variety of experimental molecular technologies used in many health sciences research laboratories, including SNP (single nucleotide polymorphism) analysis of DNA polymorphisms, RNA gene expression arrays, and proteomic platforms. In addition, the Suite facilitates the systematic management and analysis of pre-clinical, clinical, demographic, and biomedical information along with the associated molecular data in its various forms.
The analysis engine of the program is extended to include vast datasets in existing EMR systems as well as available external sources for hypothesis generation, analysis, and validation. This multi-level functionality allows users to efficiently identify statistically significant genotype-phenotype correlations within complex datasets, produce a plausible hypothesis explaining the correlations identified, and then test this and any other additional hypotheses identified in the laboratory.
Data Storage: XB Bio-Integration Suite™ stores, manages, and retrieves information and data pertaining to a subject, its derived samples, and experiments. Significantly, it can also statistically analyze these data, compile and test hypotheses, and identify viable diagnostic and therapeutic strategies for the accurate diagnoses and treatment of molecular-based diseases. The most essential element of this analysis system is that it not only archives information, but also retrieves and analyzes disparate data types within the same framework.
Data from individual subjects (for example, patients), derived samples (such as tissue, blood, urine, feces) and performed experiments (such as genotyping, cytogenetics, gene expression, proteomics) are captured through a flexible modular system that uses the metadata format to store fields and associated data. Thanks to the data storage and analysis hierocracy of XB Bio-Integration Suite™, data from an experiment is no longer painfully separated from its sample or its subject data. Rather, in a complete data set, experimental data feeds into data about the sample, which feeds into data about the subject. Accordingly, the end user is no longer relegated to analyzing data about subjects, samples, and experiments individually and then later trying to correlate them; instead the program integrates and iterates the data together for one painless data analysis set. Finally, because the strength of program is limited by the quality of data it analyzes, user entered experimental information includes quality control and assurance fields, so that possible artifacts related to the experimental procedures can readily be identified.
In addition to its unique data input system, XB Bio-Integration Suite™ includes a chronological tracking system. Assorted data types can be entered along with their associated date and time stamps such that multiple events can be tracked, associated, and plotted over the life span of the subject. Furthermore, using analysis of compiled data fields, new fields (for example, "tumor response" can be populated with data calculated from expressions of existing data within the database (for example, change in blood PSA between times t1 and t2). This represents an essential element in addressing various endpoint criteria, for example in assessing clinical response to drug X.
Multiple sample processing and associated molecular analysis are inherently cost prohibitive. Obviously, the Suite's database can be populated with internally generated data, but we have also designed a flexible template-oriented system that allows the user to import data into XB Bio-Integration Suite from the public domain. This data can be added to the user's self-populated data so that the end analysis is made even more powerful by enhancing the relevancy of clinically related studies.
Visualization Tools: A variety of proprietary visualization tools allow the user to graphically analyze the statistical correlates found between genotype and phenotype, and also biomarkers associated with the feature(s) of interest. For example, if the user is comparing normal tissue to diseased tissue, the genes/proteins that best distinguish between these two conditions are identified, and "annotated" to provide information as to which of these represent the best candidates for blood markers of the disease. Furthermore, those biomarkers which correspond to "drugable" targets for which either drugs presently exist or could be produced are also "annotated." This annotation functionality uses a series of public databases to support exploration and validation based on user defined criteria for accuracy, sensitivity, and specificity.
In line with most current academic investigations, XB Bio-Integration Suite™ can also be operated in molecular data only mode. In this case, the user can perform the standard statistical analysis (such as clustering, principle component analysis, multidimensional scaling, discrimination analysis, etc) of molecular data sets with little phenotype information (i.e., the sample number is known, but little or no information about the sample has been collected). In this instance, the ability to integrate data sets generated across different molecular platforms (for example: Affymetrix gene expression versus cDNA array gene expression versus 2D gel/Mass spectrometry protein expression) represents a feature of XB Bio-Integration Suite™ that has attracted interest from a variety of potential users.