Install dependencies

1. Make sure you have conda installed.

2. Install guppy manually.

3. Then, download nano-wgs directory and install dependencies using conda.

git -clone https://github.com/k0n5/nano-wgs.git
cd nano-wgs
conda env create -f nano-wgs-env.yaml

Run pipeline

Set up your data as follows:

<Project Directory>
├── metadata.csv
└── RawData/
    ├── <Run Directory 1>
        └── ...
            └── fast5/
                ├── <fast5 file 1>.fast5
                ├── ...
                └── <fast5 file n>.fast5
    ├── ...
    └── <Run Directory n>
        └── ...

metadata.csv is a colon-separated file specifying the different Nanopore runs, library kits, flowcell versions, barcodes and samples. Entries in the Run column have to match directory names in RawData/. There can be only one Kit and Flowcell per run. Kit and Flowcell have Make sure If no barcoding was used, use none. Samples with identical names will be merged across runs.

Example:

Run;Kit;Flowcell;Barcode;Sample
RMS_1;SQK-RBK004;FLO-MIN106;barcode01;Rh30
RMS_1;SQK-RBK004;FLO-MIN106;barcode03;Rh36
RMS_1;SQK-RBK004;FLO-MIN106;barcode04;TE441T
RMS_1;SQK-RPK004;FLO-MIN106;barcode05;T174
HMW_IMR575_run1;SQK-LSK109;FLO-MIN106;none;IMR-5-75
HMW_IMR575_run2;SQK-LSK109;FLO-MIN106;none;IMR-5-75

Then, edit the header of the <PATH TO>/nano-wgs/Snakefile to match your setup:

# PROJECT INFORMATION
PROJECT_NAME = "<CHOOSE A PROJECT NAME HERE>"
WORKING_DIR = "<PATH TO PROJECT DIRECTORY>"

# GENERAL
GUPPY_BASECALLER = "<PATH TO>/ont-guppy-cpu/bin/guppy_basecaller"
HG19 = "<PATH TO>/hg19.fa"

Once this is set up, you are ready to go:

cd <PATH TO>/nano-wgs/
conda activate nano-wgs-env
sbatch nano-wgs.snakejob.sh