GeneMaths XT: Features
1. Easy and fast wizard-driven import of data from almost all existing data sources
2. Accurate data preprocessing and normalization for all types of data
3. Efficient data selection and reduction tools through the use of layers and subsets
4. Active History concept: ability to track the different steps taken during the processing and repeat the entire analysis on other data sets
5. Appropriate data mining and analysis tools, powerful classification tools
6. Professional statistics made easy to non-experts
7. Accurate estimation of error throughout the analysis process
8. Easy and versatile plot wizard
9. Dynamic environment integrating all analysis and interface components
10. Connectivity and interactivity with external databases (Internet, pathways, etc.)
11. Automation and customization through powerful script language
1. Data import
The import of microarray data is facilitated by means of an intuitive import wizard and many predefined formats from all the major quantification programs and databases:
- Agilent Feature Extraction software
- Affymetrix (CHP-files, CEL-files, tab-delimited files)
- Arrayplot
- Arrayvision
- EBI MIAME
- Genepix
- GEO’s SOFT (raw data, datasets, and whole series)
- Imagene
- Koadarray
- ScanAlyze
- ScanArray
- Spotfinder
- …
GeneMaths XT is GeneChip compatible and Affymetrix GeneChip Command Console (AGCC) ready.
2. Data preprocessing and normalization
GeneMaths XT offers an extensive set of mathematical operations for data preprocessing and normalization:
- Background correction
- Log transformation
- Lowess normalization
- Quantile normalization
- Variance stabilization
- Imputing missing values
- …
Besides these standard normalization methods, GeneMaths XT offers the possibility to choose from a wide range of methods to preprocess Affymetrix data including MAS5.0, dChip, RMA with many variants, the PLIER method or to use a custom algorithm.
With the use of scripts, GeneMaths XT is capable of calling routines from external applications such as BioConductor.
3. The concept of layers and subsets
A Layer is one single matrix of (expression) values (and associated error values) containing all genes and experiments in the session (full X-Y matrix). Each element of the layer contains a single value together with an optional error value.
A Subset is a matrix that contains a number of rows and a number of columns from the full data matrix in a session (reduced X-Y matrix). A subset only defines the rows and columns to be included, and as such, applies to all layers of a session. The creation of subsets is extremely helpful in stepwise reducing the data matrix to genes that have an interesting expression behaviour, leaving out those that are invariant or have unacceptable errors associated with them.
4. Active History and repeatability
- GeneMaths XT constructs a dependency tree, so that the user can see the chain of actions that depend on the object that is to be modified. Based upon the tracking of dependencies, GeneMaths XT can automatically recalculate the objects and analyses in the chain if the program is requested to do so.
- GeneMaths XT also keeps track of every action done by the user, providing a global undo function. The undo chain contains all objects and analyses that have been created together with all queries and selections that were made. The full sequence of actions carried out in a session can be printed or exported as a readable RTF document that concisely describes the parameters, settings etc. used in each action or command.
- In the context of large projects, where parts of the data reduction are likely to be repeated on different data sets, GeneMaths XT allows templates to be recorded that contain a series of steps done in a mining session, and can be used on other data sets.
5. Data analysis and classification tools
GeneMaths XT offers a wide range of data analysis tools that can be used to do an analysis of groups contained in the data set without foreknowledge. They are also known as unsupervised learning methods. All clusterings and coordinate spaces are provided in an appealing and interactive interface. A useful navigator is present to switch easily between different analyses.
Data analysis tools on genes and arrays include
- hierarchical clustering
- bootstrap analysis
- Euclidean distance clustering
- partitioning
- principal components analysis
- self-organizing maps
- pattern matching
- time course analysis.
Classifiers are supervised learning methods that use group markers as input for training of the classifier. These methods use knowledge on the data set that is not contained in the expression matrix but e.g. in a text field or from external sources. Classification tools available in GeneMaths XT applicable on genes and arrays are:
- k-nearest neighbour
- neural networks
- Support Vector Machines (SVM)
6. Statistics
A variety of statistical tests are implemented in GeneMaths XT and are well documented and presented in a wizard to guide the non-expert user. Results from tests can be directly passed to the gene and array query tool. Multiple hypothesis testing procedures are also available for the Family-Wise Error Rate (FWER) control and False Discovery Rate (FDR) control.
7. The estimation of error
Error handling and error calculation is carried out throughout the workflow, and errors can be indicated numerically or graphically on any analysis report that supports the use of errors. The error values on imported layers can e.g. be derived from the spread on repeated measurements or from standard deviations on the spot quantifications.
During the preprocessing of data, a completely parallel error transformation is consequently performed. As a result, the decision of omitting data due to a bad quality indication is delayed till after the data preprocessing.
8. Graphs and plots
Plots can be derived from virtually any [combination of] data/results in the analysis, ranging from partial data selections to principal components, p-values, etc. The plot tool is wizard-driven, making it easy to generate plots of any required type.
All charts and graphs are dynamically updated when the data is manipulated.
9. Dynamic interface
Dynamic interface for microarray data analysis where different analyses can be viewed and edited simultaneously in one integrated environment. A different algorithm applied on the data is only a button click away. The following main window classes are designed according to functionality:
- Main window
- Plot window
- PCA window
- SOM window
- Partitioning window
- Time course analysis window.
The view windows of the different functionalities communicate seemlessly and can be popped up using a navigation tool. This feature optimizes the flexibility and makes the use of GeneMaths XT easier.
10. Connectivity with external databases
GeneMaths XT offers the possibility to link genes and arrays to external databases:
- Gene Ontology database
- InterPro database
- KEGG database
- Saccharomyces Genome database
- Pseudomonas Genome database
- EBI
- NCBI
- …
11. The script language
Any degree of customization can be obtained using the program's embedded script language, which allows applications and interface components to be written and used as plugins.
12. Hardware requirements
GeneMaths XT requires Pentium compatible PC running Windows XP or Windows Vista as Operating System. Memory requirement: 512 MB; 1 GB recommended. SXGA true RGB graphics required; 1200x1024 or higher resolution recommended.
Not convinced yet? Try a free demo!
|
|
Bioanalysis made easy
