-
Basecalling of sequencing data can be carried out on the DGX Station A100 using the Guppy software.
To use Guppy on the DGX Station A100, install the GPU version of the software on the station from the Debian package as described in the Guppy protocol.
-
To maximise Guppy performance on the DGX Station A100, there are two main requirements:
- Use the Guppy basecall server, so that there is a single program managing Guppy's GPU use.
- Use multiple processes to overcome issues which arise when a single process is unable to read .fast5 files fast enough.
-
This document will cover two main use cases:
- Basecalling a single folder with a lot of data in it.
- Basecalling many data folders in separate tasks, as if a job scheduler is being used.
-
At its current performance, the Guppy basecall server will occupy most of the GPU memory available, preventing its use by other programs. To enable the use of other programs such as Medaka, Megalodon and Bonito, this guide will assume the following workflow:
- Starting a new basecall server.
- Using the basecall server for basecalling, in one of the two use cases outlined above.
- Shutting down the server when basecalling is complete.