Support Nanopore raw run data delivery and GPU-based conversion in cmg-bioloop

[charlesbrandt]

Enable delivery and processing of Nanopore raw run data through the cmg-bioloop instance.

Requirements:

• Allow staging of Nanopore raw run folders (e.g., 20250903_1503_3D_PBE73759_b16fb0d9) into slate-scratch for delivery.
• For runs requiring conversions, provide a mechanism to submit GPU-enabled SLURM jobs to convert raw run signals into raw data, using user-provided job scripts (see example below).
• After conversion, enable automated ingestion of per-sample output files into the cmg-bioloop portal.
• As a fallback, support manual upload of converted raw and analyzed data.

Example SLURM job

#SBATCH -p hopper
#SBATCH -o ${SAMPLE}.out
#SBATCH -e ${SAMPLE}.err
#SBATCH --mail-type=ALL
#SBATCH --mail-user=hongao@iu.edu
#SBATCH --nodes=1
#SBATCH --ntasks-per-node=32
#SBATCH --gpus-per-node=4
#SBATCH --mem=96gb
#SBATCH --time=48:00:00
 
./exmple.job