search

bioXiaoheng / BallerMixPlus

This repository hosts the software package for BalLeRMix+, an extension of BalLeRMix that can jointly detect recent positive selection and long-term balancing selection.
MIT License
5 stars 1 forks source link

the running time was so long that i couldn't finish my task #3

Open xinkaitong opened 2 years ago

xinkaitong commented 2 years ago

Hi Xiaoheng, I have read your paper, well done! Now, the software has been installed and can be executed without mistakes. But it spent about 56 minutes executing the example file which contained 758 variants. Am I wrong? I have about 40 million variants to calculate. If so, it's impossible to finish my task.

bioXiaoheng commented 2 years ago

Hi there,

Thanks for trying out the software! Which platform/system environment are you running it on? My experience is that the example typically shouldn't take more than 30min on a regular linux machine, but if you run it on window's powershell or its linux sub-system (WSL) with conda environment, somehow python would become painstakingly slow and the task would take hours.

xinkaitong commented 2 years ago

Hi there,

Thanks for trying out the software! Which platform/system environment are you running it on? My experience is that the example typically shouldn't take more than 30min on a regular linux machine, but if you run it on window's powershell or its linux sub-system (WSL) with conda environment, somehow python would become painstakingly slow and the task would take hours.

yes, I used a regular linux machine, but it actually took me much time. Another answer I want to know is that how can I finish my task over 40 million variants using your software.

bioXiaoheng commented 2 years ago

Hi there, Thanks for trying out the software! Which platform/system environment are you running it on? My experience is that the example typically shouldn't take more than 30min on a regular linux machine, but if you run it on window's powershell or its linux sub-system (WSL) with conda environment, somehow python would become painstakingly slow and the task would take hours.

yes, I used a regular linux machine, but it actually took me much time. Another answer I want to know is that how can I finish my task over 40 million variants using your software.

You can consider parallelizing your analyses. You can find more details at issue #1