The following is a selection of frequently asked questions regarding CyTOF reagents, operation and data analysis. More questions? Contact us.
Di means 'Dual instrument' and Dd means 'Dual data'. This indicates whether the conversion to dual counts was done according to the instrument-calibrated values or data-calibrated values. Using Di is recommended both for consistency between files and for accuracy of the conversion to dual counts. If you absolutely must use Dd, make sure to adjust the 'Intensity Upper Limit Per Push' parameter to at least 500.
Most of the time, the best order is concatenate, then normalize, then debarcode. Exceptions are when there might be a discontinuity between consecutive files, such as when you use Dd mode, or if you had to adjust any settings on the instrument. In that case, if your files are large enough to contain at the very least a few hundred beads, you may want to normalize first and then concatenate. Any files with only a few beads should be concatenated first regardless of other conditions.
Somehow your FCS file (or an FCS file that makes up part of the concatenated FCS file you're looking at) did not get "randomized" from integers. Re-start the CyTOF software and convert the *_cells_found.txt files to FCS files using the 'Convert to FCS' panel. This will only take a few minutes.
The cell acquisition rate varies widely over the course of each injection. An average rate of 300 cells/second often means that the instantaneous rate varies from 0 to 700 cells/second. Increasing the instantaneous rate above 500 cells/second results in more total events per time, but often yields the same number of singlets as at 500 cells/second, hence running too fast means that there is a loss in data quality with no time gained by the user. Running at a maximum average rate of 300 cells/second, but preferably even slower, is strongly encouraged. Once the cells have been stained and are ready for the CyTOF, the acquisition rate is by far the largest influence on data quality, and running slower will certainly save time in the long term as the increased data quality simplifies the downstream analysis.