config.yaml

# Configuration file for SCVUSv3

# Author: Teresita M. Porter
# Date: February, 24, 2020

############################################################################
# How to run:
############################################################################

# Create a new conda environment (only need to do this once)
# conda env create -f environment.yml
# Activate conda environment
# conda acitivate myenv.3
# Edit filenames, paths, and parameters in this file (config.yaml)
# Run snakemake (edit number of jobs according to number of cores you have)
# snakemake --jobs 24 --snakefile snakefile --configfile config.yaml

############################################################################
# Identify raw read files
############################################################################

# This directory contains raw reads (all of them or from just one run)
# Only use compressed fastq files with this pipeline, ex. fastq.gz
# For the standard pipeline, call the directory "data"
raw: "data"

# Indicate 'sample' and 'read' wildcards in raw filenames:
# These files should be in a "data" or "run1", etc. folder
# Sample filename structure,
# 	SITE-CONDITION-REPLICATE_S1_L001_R1_001.fastq.gz
# 	{sample}_L001_R{read}_001.fastq.gz
raw_sample_read_wildcards: "data/{sample}_L001_R{read}_001.fastq.gz"

# SEQPREP sample wildcard and parameters
# These files should be in a "data" or "run1", etc. folder
# Sample,
#	{sample}_L001_R1_001.fastq.gz
raw_sample_forward_wildcard: "data/{sample}_L001_R1_001.fastq.gz"
raw_sample_reverse_wildcard: "data/{sample}_L001_R2_001.fastq.gz"

############################################################################
# Raw read pairing
############################################################################

SEQPREP:
# Phred score quality cutoff
    q: 20
# Minimum overlap length between forward and reverse reads
    o: 25

############################################################################
# Primer trimming
############################################################################

# CUTADAPT parameters
# TAReuk454FWD1 Eukaryote 18Sv4 forward primer (Stoeck et al., 2010)
# TAReukREV3 Eukaryote 18Sv4 reverse primer (Stoeck et al., 2010)
CUTADAPT_FWD:
    g: "CCAGCASCYGCGGTAATTCC"
    m: 150
    q: "20,20"
    mn: 3

CUTADAPT_REV:
    a: "TYRATCAAGAACGAAAGT"
    m: 150
    q: "20,20"
    mn: 3


############################################################################
# Dereplication
############################################################################

# Indicate a directory name here that is short and simple with no spaces or weird punctuation
# For the standard pipeline, a good directory name would be the amplicon, ex. "18Sv4"
dir: "18Sv4"

############################################################################
# Denoising
############################################################################

# Indicate minimum number of reads per cluster to retain
# Here, remove all singletons and doubletons, retain clusters with 3+ reads

VSEARCH_DENOISE:
    minsize: 3

############################################################################
# ESV x sample table
############################################################################

# VSEARCH params
VSEARCH_TABLE:
# Indicate number of threads to use
# Do not exceed the number of jobs allotted to run the whole pipeline (# jobs entered at command line above takes precedence)
    t: 24

############################################################################
# Taxonomic assignment
############################################################################

# Uses the RDP classifier
# Do not use old RDP classifier v2.2 from conda, install the newer v2.12 from SourceForge https://sourceforge.net/projects/rdp-classifier/
# 18S Classifier v3.2 based on SILVA but compatible with RDP classifier available from GitHub https://github.com/terrimporter/18SClassifier
# nuclear-encoded 18S diatom Classifier v1 based on curated barcode sequences from the INRA (Rimet et al., 2016) is available from GitHub https://github.com/terrimporter/SSUdiatomClassifier 

RDP:
    jar: "/path/to/rdp_classifier_2.12/dist/classifier.jar"
    t: "/path/to/18S_Eukaryota/v4.0/mydata/mydata_trained/rRNAClassifier.properties"

############################################################################
# Reformat CSV
############################################################################

# Add amplicon name to Zotu to keep these ids unique when data from many amplicons are combined
# The pattern will prefix the Zotu with the amplicon name
# Ex. sed -e 's/^/amplicon_/g' infile > outfile
# Below, enter the substitution pattern for sed to use (the part in single quotes above)

SED: 's/^/18Sv4_/g'