GeneSpring GX MPP 13.0 is the latest step towards quick, easy and robust expression analysis. Built on a new and more modern development platform, avadis™, from Strand Life Sciences, GeneSpring GX MPP 13.0 has gained many new features and flexibility. Frequent product updates and a persistent emphasis on making powerful statistical tools accessible through guided analysis and intuitive interface have placed GeneSpring 13.0 at the top of the list of preferred analysis applications within the community of gene expression biologists. GeneSpring 13.0 has 4400 references in Google Scholar, including over 1600 in peer reviewed publications.

Features for Version 9

Intuitive Graphical Displays
GeneSpring 13.0 displays expression data in ways that help you conceptualize the information in your data and convey it to your peers. The various types of plots, graphs and diagrams highlight different aspects of the data, allowing you to extract visual information in different ways. Virtually any graphical image can be exported as HTML or as image in .tiff, .jpeg, .png or .bmp format compatible with publishing software applications. Visualization tools in GeneSpring 13.0 include:

• 2D scatter plot
• Histogram plot
• Volcano plots
• Box-Whisker plot
• Matrix plot
• Venn Diagram
• Spreadsheet
• 2D dendrogram
• Heatmap
• Chromosome map

Zoom	Multiple visualization tools allow you to interrogate different aspects of the data. Views are linked such that selection of entities in one view automatically highlights the same entities in all other views.

Project-based hierarchical data organization
GeneSpring 13.0 data are organized in projects. Each project can contain one or more experiments, and different experiments can contain data from different array types or technologies. This organization enables users to more easily compare results from related experiments contained in the same project. For each experiment, the results of the analysis are stored in a hierarchy that maintains the order and dependency of how the results were created. This data-object hierarchy yields immediate information about how the results are related to previous results and allows user to keep track of their data analysis workflow.

Zoom	Project-base data organization enables you to store data from related experiments in a single workspace, regardless of array type or technology used. The data object hierarchy for each experiment captures information about order and dependency of how the results were generated.

Guidance for most common tasks
GeneSpring GX MPP 13.0 introduces the concept of Guided Workflows to step a user through a defined series of the most common analyses performed on expression data. For each step of the analysis, the Guided Workflow selects the most sensible choices for the majority of analysis parameters. Thus, a user only has to make a limited number of choices on their own. Guided Workflows are specific for the technology type of the data. For example, the Guided Workflow for Agilent data will use the QC metrics produced by Feature Extraction software to help assess the processing quality of your samples in GeneSpring 13.0. The Guided Workflow for Affymetrix data will use Affymetrix specific metrics, such as hybridization control plots and 3'/5' ratios. Once you have finished the Guided Workflow, all data objects will be saved to the Navigator, where they will be available for further analysis in the Advanced Analysis mode of GeneSpring 13.0. The Advanced Analysis mode gives you full control over all GeneSpring 13.0 tools and functionalities, allowing you to perform analysis in any order with any parameters for each analysis. Data of all array types and technologies can be analyzed in the Advanced Analysis mode. Guided Workflows in GeneSpring 13.0 include:

• Find differentially expressed genes for Agilent Two-color Expression Array
• Find differentially expressed genes for Affymetrix GeneChip® Expression Array
• Find differentially expressed genes for Affymetrix GeneChip® Exon 1.0ST Array
• Find differentially expressed genes for Illumina Beadchips Expression Array

Zoom	GeneSpring 13.0 incorporates vendor specific quality control metrics within an intuitive interface to facilitate the indentification of poor quality samples. These samples can then be removed from the experiment allowing you to proceed with analysis using only data from high-quality samples.

DATA Nomalization and transformation methods
Despite tightly controlled experimental conditions technical bias invariably is introduced into your data. Thus, the changes in gene expression may be attributed to true biological variation or technical varaiation. To help you answer biological questions that the experiment was designed to address. GneSrping GX provides a suite of normalizatiopn methods that wil help limit the technical variation on the data so that true biological variations are revealed. These normalization methods include:

• Probe-level and probesheet level syummarixation for Affymetrix data:
  • RMA
  • GC-RMA
  • PLIER 16
  • MAS 5
  • Li Wong
• Median/percentile shift
• Quantile normalization
• Intensity dependent (LOWESS) normalization (two-color data only)
• Dye swap (two-color data only)
• Baseline transformation to median of all samples or median on control samples

Statistical tools for testing differential expression
GeneSpring 13.0 provides a large variety of statistical tools that can be applied to test for differential expression in experiments of various designs. You can select the appropriate statistical test to apply to your dataset; examples include determining the effect of varying one or more experimental parameters; if the expression of genes is normally distributed in the population being studied; or if the variance of gene expression is equal across populations. Powerful statistical tools accessible through GeneSpring 13.0's flexible and easy-to-use interface include:

• T-tests
  • Parametric and non-parametric
  • Paired and unpaired
• One-way ANOVA
  • Parametric and non-parametric
• Multi-way ANOVA
  • Balanced and unbalanced two-way ANOVA
  • Three-way ANOVA
• Repeated Measures ANOVA
• Tukey and Student-Newman-Keuls post hoc tests o
• Multiple testing correction methods
• Permutative and asymptotic p-value Computation

Zoom	Statistical analysis result window allows you to save a list of entities that were called to be differentially expressed at the specified level of significance. Relevant information such as the calculated p-values and corrected p-values are saved along with the significant entities.

Pattern Discovery
GeneSpring 13.0 gives you a broad choice of tools for uncovering unique patterns in your expression data. Principal Component Analysis (PCA) reduses the dimensionality or complex datasets by identifying principal components that capture the variance in the dataset andy represent the most dominant profiles in hthe dataset. Clustering algorithms can be employed to group entities and/or samples based on the similarity of their expression profiles. Results from such an anlysis may reveal information regarding the biological function or co-regulation of the genes belonginf to the same cluster. In addition, GeneSpring 13.0 allows you to take an entity of interest, perhaps a gene that you know plays an important role in the disease that you are studying, and identify other genes with similar expression profiles. results from such an analysis may allow you to discover other genes that play an important role in the disease process under study. Pattern discovery tools available in GeneSpring 13.0 include:

• PCA on Entities or Conditions
• Clustering Algorithms
  Hierarchical
  • Self-organizing maps
  • k-means
  • PCS-based
• Find Similar entities Tool

Zoom

Four different clustering algorithms help identify unique expression patterns in the dataset. Entities and samples can be grouped according to the similarity of their expression profiles. An entity list for each cluster identified can be created allowing cloaser interrogation of the genes found within a cluster.

Class Prediction
GeneSpring 13.0 class prediction analysis tools use training set data to find clinically predictive patterns of gene expression data using a large variety and sophisticated algorithms that include:

• Support Vector Machines
• Neural Network
• Naive Bayesia
• Decision Tree

Biological Contextualization
One of the most critical strps of data analysis is to put statiscally significant findings into a biological context. Instead of looking at your results at the individual gene level, you cna instead explore what biological process, molecular function, cellular components or metabolic and biological pathways are significantly impacted. GeneSpring 13.0 provides you quick access to biological information for each gnee of iterest. Biological relevance of your statistical resultss can also be obtained by determining whether there is a significant enrichment of your genes of interest in any pathways of Gnee Ontology terms. Biological contextualization tools provided in GeneSpring 13.0 include:

• GO Analysis
  • View functional classification of genes within a GO Browser
  • GO enrichment analysis
• Pathway Analysis
  • Import andd view BioPAX pathway exchange format (OWL)
  • Find Similar Pathways tool
• Gene Set Enrichment Analysis
  • Use of Gene Sites from the Broad Institurte for GSEA
  • Use of any list of genes for GSEA