|
|
Comparison tools
The Plugin Tools offer a wide variety of additional useful tools to the BioNumerics software, provided as a service by Applied Maths.
The plugin tools are based on the powerful BioNumerics script language, which makes it possible for the user to customize them according to personal needs. They can be run directly from the Applied Maths website, or can be downloaded to the local BioNumerics script folder. To download scripts, press the  in the button toolbar of this window.
You are free to run and/or download these tools for personal use, and modify the scripts as needed. Redistribution or reproduction of the plugin tools by any means is prohibited.
Each plugin tool is provided "as is" and with no further liability or guarantee from Applied Maths. Any consequences that may arise from the use of these tools are at your own responsibility.
 Auto collapse branches
This tool can be used to have the branches of a tree automatically displayed as "abridged". The minimum similarity has to be entered as well as the minimum number of entries a branch should contain.
Export tree as text
With this tool, you can export a tree in text format, using the so-called "Newick" format. The script asks to select a database fields to label the branch tips, and th enter a file name and path for the exported text file.
Export tree as XML
With this tool, you can export a tree in XML format, using the so-called "Newick" format. The script asks to select a database fields to label the branch tips. You can also have the XML files exported with a hierarchical node set or a flat node set. You can enter a file name and path for the exported XML file.
Find exemplars
Based upon the similarity matrix, this script will define an exemplar for each group you have defined in a Comparison. The exemplar is the member of a group that has the highest average similarity with all the other members in the group.
The script first asks to specify an experiment type, which can be of any class. When a similarity matrix is available for the selected experiment type, it pops up a report listing the exemplars found and their average similarity with the other group members. The exemplars are selected in the comparison window. Warning: any previous selection made will be cleared by this script.
Create centrotypes
This script will create a centrotype for each group you have defined in a Comparison. The centrotype is the "average" member of a group, i.e. having the average value for each character. As such, it is a theoretical profile which is not present in the data set. It is obvious that this tool only applies to non-binary character data.
The script first asks to specify an experiment type, which should be a non-binary character type. It pops up a report listing the centrotypes and their profiles. Using the button <Add to database>, you can add the centrotypes as new entries to the database. The script will allow you to choose a database field to enter a string, e.g. "Centrotype". When created, the centrotypes are automatically assigned to the groups from which they were created.
Map protein alignment on DNA sequences
The script is designed to allow protein encoding sequences to be aligned at amino acid level, of which the multiple alignment is mapped back on the DNA sequences. While performing a multiple alignment on protein sequences allows easier and more reliable multiple alignments to be created, performing the cluster analysis on the correspondingly aligned DNA sequences gives more sensitive distance estimations.
The tool maps the protein alignment onto the corresponding DNA sequences. This plugin tool relies on the existence of a nucleic acid Sequence type and a corresponding translated amino acid Sequence type which has been created by the plugin tool Translate selected sequences under "Sequence related tools", which should therefore be executed first.
To execute the alignment mapping for a Sequence type *SEQTYPE*, open the corresponding protein sequence type *SEQTYPE*_TRANSL in the Comparison window, and perform the multiple alignment. Open the DNA Sequence type *SEQTYPE* and run the script. The sequence type *SEQTYPE* will contain the aligned DNA sequences, which can be clustered.
Note: This script installs itself as a menu item Sequence > Map protein alignment on DNA sequences when copied to the local BioNumerics script folder.
Cutoff clusters below a threshold
Below a similarity level you specify, clusters will be displayed in dashed lines. A minimum number of members can also be specified, above which clusters will be displayed in full lines. This function can be useful for printing and publication purposes, to distinguish insignificant clusters from significant clusters.
Export aligned sequences
Allows aligned sequences to be exported from a comparison. Gaps are presented as spaces. The difference between this tool and the menu function "File > Export sequences" in the comparison window, is that you can choose what fields to export.
Fill field with cluster number
You can specify a similarity value, at which level the dendrogram is isplit up into clusters. The thus obtained clusters are numbered, and the cluster numbers are filled into a field you specify. You can also specify the minimum number of entries a cluster should have in order to be counted. By checking "Save fields", the field containing the cluster number is permanently saved.
Find entry in comparison
Within a comparison, you can enter a search string and specify any of the available fields or all fields at once. The asterisk (*) wildcard can be used to enter incomplete strings. With the "Regular expression" checkbox, you can enter a regular expression, which adds power to the search capacity (see the manual about regular expressions).
Plot number of clusters versus similarity
This tool displays a graphical X-Y plot with the cumulative number of clusters plotted against the similarity scale. The plot gives an idea about the range of similarity most clusters fall in. The plot can be copied to the clipboard as enhanced metafile, or printed.
Find best approximating branch
The tool can be used to compare two dendrograms in a comparison. One can select a branch in one of the dendrograms and the script will find a best matching branch in another dendrogram.
To operate the script, first select a branch on the dendrogram to start from. To make interpretation of the result easier, you can select the entriers of that branch (CTRL + left mouse click on the branch). Then launch the script, which will ask you to specify the experiment type of the second dendrogram. In a next dialog box, non-exactly matching branches can be dealt with by allowing relative penalties (in percentage) to be set for entries present in the branch of the first tree but not in the branch of the second tree, and vice versa.
After the script has been run, the second tree is displayed, and the best approximating branch is selected on it.
If downloaded to the local computer, the script inserts itself as a menu item Clustering > Find best approximating branch.
Rearrange dendrogram to match another dendrogram
While comparing two dendrograms resulting from different techniques, this tool will facilitate the interpretation and comparison of both dendrograms, by rearranging the branches of the current dendrogram so that the order of entries approaches the order of entries in the other dendrogram as closely as possible. When the tool is run, it asks for the other experiment that will serve as a "standard" to rearrange the tree. Al list of all available experiments is shown. Obviously, the rearrangement can only be done when a clustering is present for the two experiments.
Remove all groups
With this function, you can clear all existing groups in a comparison. It replaces selecting all the entries from the comparison, and clicking "Groups" > "Assign selected to" > "None".
Show group legend
With this function, you can create a legend of the group colors and the corresponding names, chosen from a specific field.
Search for same cluster in other dendrogram
You can select a cluster in a dendrogram resulting from one experiment, and this tool will search for the closest matching cluster in a dendrogram resulting from another experiment. A list of the available other experiments is given from which you can select. In addition, one can specify the cost for extra entries found in the matching cluster. When 100% is specified, an extra entry is penalized as heavily as a missing entry. When 0% is specified, no cost is given to extra entries, and the root of the other dendrogram will always be selected. After execution of the script, the entries of the chosen branch are selected (blue arrows), and the dendrogram of the other experiment is displayed, with the closest matching branch selected. Note that any other selection of entries you made in the comparison will get lost by this script.
Display sequence diversity
In a comparison with aligned sequences, this tool will calculate the percentage occurrence and degree of consensus of each position in the multiple alignment. The result is presented graphically, with gray bars showing the degree of consensus of each position, and black dots the relative occurrence. The degree of consensus is calculated as [(nA²+nC²+nG²+nT²)/n²-0.25]/0.75, which results in a number between 0 (four bases each 25%) and 100 (same base for all sequences).
Simpson's Index of Diversity
In a comparison with groups defined, this tool will calculate Simpson's Index of Diversity. It returns the total number of entries, the number of entries in each group, and the Index of Diversity.
Shannon-Weiner Index of Diversity
In a comparison with groups defined, this tool will calculate Shannon-Weiner's Index of Diversity. It returns the total number of entries, the number of entries in each group, and the Index of Diversity, based on natural logarithms.
|
