Known bugs, problems and grievances

DescriptionVersion
The "default values" of the Sequence Numeric Maps produced by the statistic operation cannot be trusted. Version 2
( Fixed in v2.0.-1 )
( Applies not to v1 )
Sometimes when executing a command which replaces an existing data object, the object in question is not properly updated. However, if one deletes the original object first and then executes a command to create it anew it works just fine. The behaviour is inconsistent and not reproducible between sessions, so it is difficult to pinpoint the exact cause of this problem. . Version ?
A serious bug was discovered pertaining to local caching of data from region tracks. Read more. Before v1.08
The manual "edit region" functionality in the visualization panel (invoked by double-clicking on a region while holding down CONTROL) and "delete region" functionality (invoked by pointing at a region and pressing DELETE) does not work properly for module regions. This also includes UNDO/REDO of these actions.Version 1.x
Some installations of MotifLab may contain a faulty datatrack configuration file which makes it impossible to correctly import DNA sequence data for the hg19 genome build with the "Add Feature Data Tracks" function. Rather than importing the proper sequence, all the bases are set to N's (importing directly from FASTA files should work fine, however). To fix this issue, go to the "Configure" menu and select "Configure Datatracks". Next, press the "Configure Datatracks and Sources" button and edit the "DNA" track. Select the line in the data sources table which reads "Human, hg19, GET, ,tare.medisin.ntnu.no" (with a blank Format column) and press the "Edit Source" button. Finally, select "FASTA" from the Data Format drop-down menu. After saving the new configuration, you should probably bring up the Options dialog (under the Configure menu) and clear the cached feature data. A short video describing this procedure can be found here
Deleting temporary files (of which there sometimes can be quite many) after MotifLab exits can take a long time and hog a lot of CPU resources, especially on computers running Windows.Version 1.x

(not yet fixed in v2.0)
For most motif discovery tools that have the option to search either both strands or a single strand (such as ChIPMunk), the single strand option will only search the direct strand of the sequence when its orientation is known. Future releases of MotifLab will probably search the relative strand instead (i.e. the DNA sequence from the strand that the sequence is assigned to).Version 1.x

(not yet fixed in v2.0)
When creating output of Region datasets in EvidenceGFF format, region properties will not be properly output if the property name contains uppercase letters. Version 1.x



Report bug

MotifLab is a large program and will probably still include a few undetected mistakes and problems.
If you find something you believe to be an error in the program, please submit a description of the problem using the form below (try to be as specific as possible). Also, please include your e-mail address in case we need to contact you for follow-up questions.