What kinds of info indicate a datapoint is invalid?
Technical Support and Questions
Anne Stanton
Jun 2015
Hi, I'm trying to sort through my dataset and determine which datapoints are useful and which ones ought to be discarded (probably due to operator error). At this point I question the NPQt values that show up as negative. Can these be disregarded and the rest of that datapoint kept or is the whole measurement suspect? Is there tutorial for how to sift through the data and make these kinds of decisions? Thanks, Anne
Greg Austic
Jul 2015
Anne you and I had a long discussion on this, just wanted to post here to say that the answer is long and kind of complicated and requires a full video which I hope to make soon -
Cindee Giffen
Mar 2019
Hi Greg - I'm wondering if that video ever was made, or if there was a comprehensive list of errors one could refer to?
Thanks, Cindee
David M. Kramer
Mar 2019
Hi Cindee,
The error detection is set in the macro for a protocol. Someone who makes a macro can thus adjust what would be flagged as an error or a potential problem.
If you tell me what protocol you are using, I can help you determine what is an "error." But, in the meantime, there are some obvious errors, such as Nan values (not numbers) or SPAD values out of a certain range, or values for fluorescence below what one would expect, e.g. Phi2>1 or Phi2<0 would probably be flagged. One can also add flags to other kinds of potential issues by adjusting the macro.
Hope this is helpful!
Dave