Hi folks,
Does anyone have experience recovering runs that fail mid-cherrypicking? I was hoping that running our script as a method would save the temp worklist and we could continue from where the crash/error occurred, but that doesn’t seem to work.
Has anyone played with passing a count variable in the worklist to try and keep a tally of wells that were successfully completed? Any suggestions are appreciated!
Thanks!
I have one thing that might work, but it might not be an elegant solution and I am not sure yet how to implement this yet.
Some time ago I tried to access the value of a specific variable inside a liquid class, but of course the ‘standard’ that I normally use would not go inside that scripting part.
Instead I did the following:
- made a ‘set variable’ with a specified file name
- add a function to export the variable to the file name
- after the pipetting action I was able to open the file and extract the variable that was used
In your case you would do the same and place the action after each dispense
- create file name with ‘date() + something else.’ (.csv or .txt)
- unique enough for a specific day, but not to unique that it would write different files for every minute
- Export a variable to the file
- you can just count the number of lines in the file to see what has been done, although this might be tricky as it does not always pipette in the 1,2,3,4,5,6,7,8 FCA order you might expect.
- This could be circumvented if you add tip number or volume or something else unique that is used in the liquid class that could help you out with retrieving the information later on when needed.
I think that this could work, but have not tested it. Hope it nudges you in the correct direction
2 Likes
@dirkvanessen Thank you so much for this suggestion! I need to mull it over a little, but that’s an interesting idea. I hadn’t thought about that the tips won’t necessarily be used in the expected order, I could probably include a tipmask in the worklist to force it to use tips as I’d like…I think? I haven’t needed to mess with tip masks before. Appreciate your response, I wasn’t thinking to do this within the liquid class and my optimism is restored!
One thing I do is place comments in my liquid classes so I can review them in the AuditTrail when something goes wrong.
There must be a way to comment the source and destination labware & well within the microscript. I can reach out to Tecan and ask if there’s a strategy to query this information from within the microscript.
Alternatively, I’ve seen some folks break up their .GWL’s so it’s 8 wells at a time.
1 Like
Hi Luis,
Yes, you can comment the labware and well in the microscript. The variables you are looking for are “labwareName” and “indexWell”. These are automatically set for each instance of the liquid class that is generated during a run. The variable “indexTip” will tell you the current tip being used to pipette that sample. These variables are available in both the Aspirate and Dispense sections of the liquid class.
You can also use these variables to pull in other attributes from the labware that might be useful to have in your microscript via the GetAttribute() function. Alternatively, you could also set labware attributes (SetAttribute() function) in the microscript which would then allow for information to be accessed later in your script outside of the current pipetting command.
Hope this helps. -Eric