| 1. General |
| 1.1. | Why won't the program process the data file that was just read in? |
| 1.2. | Why won't the program start? |
| 1.3. | How do I cite GeneCluster? |
| 2. Marker Analysis/Neighborhood Analysis Window |
| 2.1. | Near the top of the window there appears a warning "MISMATCHED TRAIN SET". What is a mismatched train set? |
| 2.2. | Running the algorithm opens an "Error Messages" window that displays the error: "Internal error: java.lang.ArrayIndexOutOfBoundsException". What went wrong? |
| 3. Build Predictor Window |
| 3.1. | Why doesn't running the algorithm produce any predictors? |
| 3.2. | Near the top of the window there appears a warning "MISMATCHED DATASET". What is a mismatched dataset? |
| 3.3. | Running the algorithms causes the "Error Messages" window to open and displays "Internal error: java.lang.ArrayIndexOutOfBoundsException". What went wrong? |
| 3.4. | The Fisher test does not always produce results. |
| 4. Class Finder Window |
| 4.1. | Why does running this algorithm take such a long time to complete, even when running with only a few iterations? |
|
1. General |
| 1.1. | Why won't the program process the data file that was just read in? |
| | This can occur for several reasons. The most likely problem is that the dataset was not transformed. From the main window's menu choose Dataset -> Preprocess. The Dataset Preprocessing window will appear. Choose the dataset that was just loaded and add a transpose filter. Click on the "Apply and Close" button and try processing the newly created dataset. |
| 1.2. | Why won't the program start? |
| |
The program requires Java 1.3.1 or 1.4. Earlier versions of the program, especially Java 1.1 and 1.0, will not work at all. Versions 1.2 and 1.3.0 may be able to start the program, but there will be bugs and some features will fail completely.
If you are running Java 1.4.0 and you have a network drive or other device that has become inaccessible (if your machine was disconnected from the network, for example), the application will patently wait for the networked drive to become available, which will cause the program to freeze. This problem does not seem to occur in Java 1.3.1. The current solution to this problem is to disconnect all unavailable network drives.
|
| 1.3. | How do I cite GeneCluster? |
| |
You can cite this publication regarding the code:
Reich, M., Ohm, K., Tamayo, P., Angelo, M., Mesirov, J.P. (2004) GeneCluster 2.0: An advanced toolset for bioarray analysis. Bioinformatics 2004; doi:10.1093/bioinformatics/bth138
and/or these two publications regarding the basic algorithms:
Tamayo, P., Slonim, D., Mesirov, J., Zhu, Q., Dmitrovsky, E., Lander,
E.S., Golub, T.R. (1999) Interpreting gene expression with
self-organizing maps: Methods and application to hematopoeitic
differentiation. Proc. Natl. Acad. Sci. USA 96:2907-2912, 1999.
Golub, T.R., Slonim, D.K., Tamayo, P., Huard, C., Gaasenbeek, M.,
Mesirov, J.P., Coller H., Loh, M.L., Downing, J.R., Caligiuri, M.A.,
Bloomfield, C.D., Lander, E.S. (1999) Molecular classification of
cancer: Class discovery and class prediction by gene expression
monitoring. Science 286:531-537, 1999..
|
|
2. Marker Analysis/Neighborhood Analysis Window |
Discrete Metrics tab |
| 2.1. | Near the top of the window there appears a warning "MISMATCHED TRAIN SET". What is a mismatched train set? |
| | The number of data in the template must match the dataset's number of rows. The template was made for a different dataset, the dataset needs transposing (see above), or there are no templates (cls files) loaded. |
Continuous Metrics tab |
| 2.2. | Running the algorithm opens an "Error Messages" window that displays the error: "Internal error: java.lang.ArrayIndexOutOfBoundsException". What went wrong? |
| | The dataset needs to be transposed. (See above) |
|
3. Build Predictor Window |
| 3.1. | Why doesn't running the algorithm produce any predictors? |
| | In the "Generation" panel the "Build" tab must be chosen instead of "Validate". |
| 3.2. | Near the top of the window there appears a warning "MISMATCHED DATASET". What is a mismatched dataset? |
| | The number of data in the template must match the dataset's number of rows. See a similar discussion in the "Marker Selection/Neighborhood Analysis Window" section. |
| 3.3. | Running the algorithms causes the "Error Messages" window to open and displays "Internal error: java.lang.ArrayIndexOutOfBoundsException". What went wrong? |
| | In the "Inputs" panel the "Ratio" option under "Preprocess" is broken when the "Feature Selection" panel is set to the "All Pairs" option for "Binary Distinction". |
| 3.4. | The Fisher test does not always produce results. |
| | The Fisher Test will only work for two class predictors. |
|
4. Class Finder Window |
| 4.1. | Why does running this algorithm take such a long time to complete, even when running with only a few iterations? |
| | The dataset needs to be transposed. See the general discussion section. |
