Entering edit mode
10 days ago
Akshay
•
0
I am currently performing WGS using ONT P2 Solo. I noticed that the basecalling % gets lower and is stuck at 26.44 Gb while the estimated bases increases (as expected as it is real time). I am assuming this is an issue with GPU not performing live base calling? Pretty weird seeing that my previous run below had 100% basecalled I have a core i7 14th gen (28 threads), 64GB DDR5 and RTX4090. When I look at the task manager the GPU usage is low. What is the issue here? And is a potential solution to rather basecall the pod5 file post-sequencing to give better results?
Did this block eliminate itself over time? It looks like ~4 M initial reads in your run are in "paused" state. Have not heard of this happening before.
Yikes. This is scary. I'd definitely contact ONT.
No other error messages? Regarding this:
Right now at least I cannot think of any other solution ..
I've heard of cases in which the GPU gets overloaded and quits, though this was associated with the loss of the run in those cases ... not sure why this is happening but what is happening looks clear enough - after a certain point in time, you quit calling anything new. So from that time point, but the red and green flatline, though your total read keeps going up, which is encouraging because at least you have the squiggles (i hope).