How do I adjust ArrayMiner's column order to match GeneSpring's ?

    Some GeneSpring versions contain a glitch in the External Program Interface that causes ArrayMiner to receive the data with columns in the wrong order, or to receive more columns than are defined by GeneSpring's Experiment Interpretation (please refer to GeneSpring documentation). This is however very easily corrected by following the procedure below.

In GeneSpring
 
In ArrayMiner
1 Run ArrayMiner using the External Program Interface    
 
   
 
2 Verify the column order by reviewing the "Attribute" column in the Raw Data window
    3 Switch back to the running GeneSpring
   
 
4 Invoke the spreadsheet view of your data by selecting "View / View as Spreadsheet" option in GeneSpring's main menu. Check the "Show normalized" checkbox only, select the first row in the spreadsheet by clicking the name of the first gene in the table, and click the "Copy Selected" button (as shown in the figure below). This copies the column order into your machine's Clipboard.You may now close the spreadsheet window in GeneSpring.    
 
   
    5 Get back to the running ArrayMiner, and select the "Tools / Get GeneSpring data order from Clipboard" option in the main menu. Click "Yes" in the message box "Paste GeneSpring columns from Clipboard."


Copying GeneSpring's column order into the Clipboard

    This will adjust the columns in ArrayMiner to appear in the same way as in GeneSpring. If necessary, the data are then reloaded to reflect the modified column order.

NOTE

You will need to do this only once for each of your data sets. ArrayMiner keeps track of the data and automatically adjusts the columns on data you have already worked on.


NOTE

The best time to adjust the columns is before the start of a heavy clustering computation. In case your columns need adjusting, your data will be reloaded, discarding any current clustering result.