ejongepier / nanoclass

NanoClass

NanoClass is a taxonomic meta-classifier for 16S/18S amplicon sequencing data generated with the Oxford Nanopore MinION. With a single command, you can run upto 11 popular classification tools on multiple samples in parallel, including BLASTN, Centrifuge, Kraken2, IDTAXA, MegaBLAST, dcMegaBLAST, Minimap2, Mothur, QIIME2, RDP and SPINGO. Optional read preparation steps, such as quality trimming, length filtering and sub-sampling, are an integral part of the pipeline.

NanoClass automatically installs all software packages and dependencies, downloads and builds required taxonomic databases and runs the analysis on your samples.

Getting started

Requirements

The entire NanoClass workflow can be run on a powerfull desktop computer, but for many applications a laptop will do. Most classification tools implemented in NanoClass run in a matter of minutes to hours. Prerequisites are Conda and Snakemake.

NanoClass automatically installs all other software packages and dependencies.

Installation

You can either clone NanoClass, like so:

git clone https://github.com/ejongepier/NanoClass.git

Or download and extract the zip archive from https://github.com/ejongepier/NanoClass.

NanoClass is immediately ready for use. See also the Documentation.

Usage

Quick start

Enter your samples and the paths to your fastq.gz files in the sample.csv. Sample labels should be unique. Both sample and run labels should contain letters and numbers only. Barcode column should be left empty, meaning your input files should already be demultiplexed. For an example see the sample.csv file.

After editing the samples.csv, the entire pipeline can be run with a single command:

snakemake --use-conda --cores <ncores>

Where --cores are the number of CPU cores/jobs that can be run in parallel on your system.

Customizing

You can customize the pipeline through the config.yaml, e.g. by running only a subset of the reads or classification tools, or by changing the default Silva 16S taxonomic database.
For details on how to customize NanoClass, see the Documentation.

Report

After successful execution, you can create an interactive HTML report with:

snakemake --report report/NanoClass.zip

Authors

• Evelien Jongepier ([email protected])

nanoclass's Issues

Classification consensus sequence file.

Hi,

I am conducting metabarcoding on eukaryotic marker genes (e.g. nuclear rRNA clusters and mitochondrial genes).
Threfore, I understand that I have to get my sequence databases of interest in place in the correct directory.

However, I would like to compare the classified reads (clusters) directly as well to the respective databases.

Where can I find the consensus sequence per OTU? Or is the classification of the reads only basedon matches with the respective databases?

kind regards,

T.

edit adapters.py

Hi,

where can the adapters.py file of porechop be edited?
Since the env is created on the fly by snakemake it is unclear where I can edit this file.

kind regards,

Thierry

Total analysis time

Hi,

since last week I have a Nanoclass 16S analysis running (approximately 3 Mreads on 4 samples) on 56 cores. I have selected the blastn method.
The last update of the snakemake output/log mentioned 48% since last Wednesday.

I am doubting this is the way Nanoclass was meant to be used, Is there something that I overlooked?

kind regards and thanks in advance,

Thierry

AttributeError: 'str' object has no attribute 'name'

I am returning below error when running NanoClass.

Can you please help to resolve it?

Missing input files for rule common_plot_tax

Hello Evlien,

with a fresh install from github, a new samples.csv file and the UNITE db in db/common/ I get the following error just after startup

Building DAG of jobs...
MissingInputException in line 33 of /media/minion/Data/Analyses/FlevoRUN/NanoClass/rules/common.smk:
Missing input files for rule common_plot_tax:
classifications/FlevoRUN3/blastn/PLA_RON_t=1_A.blastn.taxmat

It seems to lack a link to the taxmat files.

kind regards,

Thierry

wget Silva

Hello,

wget of the Silva database comes with an error (nowadays?).
I have circumvented it by introducing the --no-check-certificate option.
Is this a proper option? If so, could you include it in the code.

Kind regards,

Thierry

Error in generating report

Hi,

after a flawless run of a metabarcoding dataset on the UNITE db I wanted to generate the report.

snakemake --report report/NanoClass.zip

Building DAG of jobs...
Creating report...
Adding dummymost1.filtered.pdf (0.091 MB).
Adding dummymost2.filtered.pdf (0.094 MB).
Adding dummymost1adapted.filtered.pdf (0.076 MB).
Adding dummymost2adapted.filtered.pdf (0.072 MB).
Adding aabund-Phylum-by-sample.pdf (0.0056 MB).
Adding rabund-Phylum-by-sample.pdf (0.0057 MB).
Adding aabund-Class-by-sample.pdf (0.006 MB).
Adding rabund-Class-by-sample.pdf (0.006 MB).
Adding aabund-Order-by-sample.pdf (0.0062 MB).
Adding rabund-Order-by-sample.pdf (0.0063 MB).
Adding aabund-Family-by-sample.pdf (0.0066 MB).
Adding rabund-Family-by-sample.pdf (0.0065 MB).
Adding aabund-Genus-by-sample.pdf (0.006 MB).
Adding rabund-Genus-by-sample.pdf (0.0062 MB).
Adding runtime-by-sample.pdf (0.0051 MB).
Adding runtime_log-by-sample.pdf (0.005 MB).
WorkflowError:
Error loading caption file of output marked for report.
WorkflowError:
Failed to open source file /home/minion/git/NanoClass/rules/../report/fig-nanofilt.rst
FileNotFoundError: [Errno 2] No such file or directory: '/home/minion/git/NanoClass/rules/../report/fig-nanofilt.rst'

Kind regards,

Thierry