Config

#!/bin/bash

#   Name this project
PROJECT="lettuce_dev"

#   What email should we use for job notifications?
EMAIL="ely67071@uga.edu"

#	Is data paired-end? ("True" or "False")
PE=True

#   Are you submitting the job with qsub from PBS (Portable Batch System) or with sbatch from Slurm?
#       Choose from: "PBS" or "Slurm"
QUEUE=Slurm

############################################
##########  Reference Genome Info ##########
############################################

####The following variables are needed for: Genome indexing, and Reference Prep

#	What file format is your genome annotation file? (GTF or GFF3)
#	This variable will be used in the 'Genome Index' and 'Reference Prep' steps
ANNOTATION_FORMAT="GTF"

#   Where is the Genome Annotation file (GFF3 or GTF)?
#   Include the full filepath
#	This variable will be used in the 'Genome Index' and 'Reference Prep' steps
GEN_ANN="/scratch/ely67071/lettuce_dev_data/genomes/Hildegardhap1v1_2.gtf"

#   Where is the Genome FASTA file?
#	Include the full filepath
#	This variable will be used in the 'Genome Index' and 'Reference Prep' steps
GEN_FASTA="/scratch/ely67071/lettuce_dev_data/genomes/Hildegardhap1v1.fasta"

############################################
##########   Quality_Assessment   ##########
############################################

#   QUEUE=PBS: What are our QSub settings for Quality_Assessment?
#       Below are the recommended settings
QA_QSUB="mem=1gb,nodes=1:ppn=4,walltime=12:00:00"

#   QUEUE=Slurm: What are our sbatch settings?
#       Below are the recommended settings

PARTITION="batch"
QA_SBATCH="--nodes=1 --ntasks=4 --mem=5gb --time=06:00:00 --mail-type=BEGIN,END,FAIL --mail-user=${EMAIL} --partition=${PARTITION}"

#   What is the file name suffix for files to be assessed?
SUFFIX="fq.gz"

#	Where is the directory containing your data to be assessed? 
#	This can be a directory to multiple directories each containing a forward and reverse read which is the format of raw data from the GGBC
#       Include the full file path, e.g. path/to/directory (with no trailing forward slash)
QA_INPUTDIR="/scratch/ely67071/lettuce_dev_data/raw_rna_seq"

#   Where do you want the FastQC results?
#       Include the full file path, e.g. path/to/directory (with no trailing forward slash)
QA_OUTPUTDIR="/scratch/ely67071/lettuce_dev_data/raw_rna_seq/quality"

#	A place for FASTQC to store temporary files
QA_TEMP="/scratch/ely67071/temp"

############################################
##########    Adapter_Trimming    ##########
############################################

#   QUEUE=PBS: What are our QSub settings for Adapter_Trimming?
#   Below are the recommended settings
AT_QSUB="mem=10gb,nodes=1:ppn=4,walltime=12:00:00"

#   QUEUE=Slurm: What are our sbatch settings?
#       Below are the recommended settings
AT_PARTITION="batch"
AT_SBATCH="--nodes=1 --ntasks=4 --mem=13gb --time=12:00:00 --mail-type=BEGIN,END,FAIL --mail-user=${EMAIL} --partition=${AT_PARTITION}"

#   What files to trim?
#	This handler will accept either a list of forward samples OR
#	a directory to raw data- this can be a directory to multiple directories,
#	each containing a forward and reverse read (such as the format of raw data from the GGBC)

#	Either a text file containing a list of samples to trim (if PE, only need to list the forward reads) including the full file path, 
#	OR a directory of input files
AT_INPUT="/scratch/ely67071/sunflower_dev_data/raw_rna_seq"

#   Where do you want the Trimmed Samples to go?
#       Include the full file path, e.g. path/to/directory (with no trailing forward slash)
AT_OUTPUTDIR="/scratch/ely67071/sunflower_dev_data/trimmed_reads"

#	What is our adapter file? Include the full file path.
ADAPTERFILE="/scratch/ely67071/sunflower_dev_data/illumina_adapters.txt"

#       File name/path of where JOB IDs should be recorded
AT_JOB_LOG="/scratch/ely67071/sunflower_dev_data/AT_job_log.txt"

#   What shared suffix do the forward samples have?
#       Example: _1_sequence.txt.gz
FORWARD_NAMING=R1_001.fastq.gz

#   What shared suffix do the reverse samples have?
#       Example: _2_sequence.txt.gz
#	(Only relevant for paired-end data)
REVERSE_NAMING=R2_001.fastq.gz

#	The maximum mismatch count allowed for the "seed" (small section of adapter),
#	which causes the entire alignment between the read and adapter to be scored.
SEEDMISMATCH=2

#	The minimum alignment score threshold for clipping adapter sequence
#	A palindroma approach is used to check for adapter 'read-through'
#	This strategy is only used in PE data, but a value must still be supplied if SE data
PALINDROMECLIP=30

#	The minimum alignment score threshold for clipping adapter sequence
SIMPLECLIP=10

#	The minimum length of adapter sequence to be removed
#	Only relevant for Paired-end data
MINADAPTERLEN=1

#	Whether to keep the reverse read if adapter read-through has been detected by palindrome mode
#	the default behavior is to entirely drop the reverse read
#	Only relevant for Paired-end data
KEEPREADS=true

#	Low quality bases are removed from the beginning of the sequence.
#	What is the minimum quality value required to keep a base at the beginning?
LEADCUT=3

#	Low quality bases are removed from the end of the sequence.
#	What is the minimum quality value required to keep a base at the end?
TRAILCUT=3

#	Reads below a specified minimum length are removed (dropped after other processing steps)
#	What is the minimum length required of reads to be kept?
MINLENGTH=20

############################################
##########    Genome_Index        ##########
############################################

#####IMPORTANT:
# The variables: 
# "ANNOTATION_FORMAT", "GEN_ANN" and "GEN_FASTA"
# MUST also be supplied (above, under 'reference genome info')

#   What are our QSub settings for Genome_Index? 
#	note that STAR needs at least 30gb of memory to run
#   Below are the recommended settings
GI_QSUB="mem=50gb,nodes=1:ppn=4,walltime=450:00:00"


#   QUEUE=Slurm: What are our sbatch settings?
#       Below are the recommended settings
PARTITION="batch"
GI_SBATCH="--nodes=1 --ntasks=4 --mem=80gb --time=96:00:00 --mail-type=BEGIN,END,FAIL --mail-user=${EMAIL} --partition=${PARTITION}"

#	Define the number of threads to be used. 
#	This must be set to the number of available cores on the server node
NTHREAD=4

#   Where do you want the files for your genome index?
#	This direcotry has to be created before program will run and needs writing permissions
#	This directory path will also be used in the next read mapping step
#	Include the full filepath
GEN_DIR="/scratch/ely67071/lettuce_dev_data/genomes/genome_indices_gtf"

#   Specify the length of genomic sequence to be used in constructing the splice junctions database.
#   This length should be equal to ReadLength-1, where ReadLength is the length of reads
#	(ex. for 2x100bp paired-end reads, the ideal value is 99)
SPLICE_JUN=149

#	What tag name should be used as the exons' gene-parents (in GTF files this defaults to "gene_id")
#	You only need to fill this out if: 1.) you are used a .gff3 file for annotations AND 
#										2.) you want GeneCounts output from read mapping
#	(The tag name used for transcript-parents for a .gff3 file is 'Parent')
#	This variable will be ignored if annotation file is GTF format
GENE_PARENT="Parent"

############################################
########    Collect_Junctions        #######
############################################

### (Optional)
### Run a first-pass mapping step with STAR to detect novel splice junctions
### These novel splice junctions can be used to regenerate the genome index for better mapping sensitivity

###### IMPORTANT: This step shares variables with Read-Mapping (below)

#   Where do you want your output files to go? Include the full file path
CJ_OUTPUTDIR="/scratch/ely67071/lettuce_dev_data/STAR_output_gtf/junction_data"
#       File name/path of where JOB IDs should be recorded
CJ_JOB_LOG="/scratch/ely67071/lettuce_dev_data/CJ_job_log.txt"

############################################
########    Filter_Junctions        ########
############################################

### Filter identified junctions for accuracy
### Outputs a filtered list that can be added to Read Mapping to improve alignment

#	What are our QSub settings for Filter_Junctions?
#	Below are the recommended settings
FJ_QSUB="mem=22gb,nodes=1:ppn=4,walltime=4:00:00"


#   QUEUE=Slurm: What are our sbatch settings?
#       Below are the recommended settings
PARTITION="batch"
FJ_SBATCH="--nodes=1 --ntasks=4 --mem=22gb --time=04:00:00 --mail-type=BEGIN,END,FAIL --mail-user=${EMAIL} --partition=${PARTITION}"


#	Where are the "SJ.out.tab" files? Include the full filepath
#	This will also be where the concatenated filtered list is put
JUNCTIONDIR="/scratch/ely67071/lettuce_dev_data/STAR_output_gtf/junction_data"

#	Name of final list (will end in _SJ.filtered.tab)
SJ_LISTNAME="lettuce_dev_"

## FILTERS:

#  Remove junctions identified in scaffold sequences?
### If yes, what string denotes non-chromosomal sequences, eg "Chr00"? (Otherwise, put NA)
SCAFFOLD_STRING="NA"

### Filter out non-canonical junctions? (yes or no)
REMOVE_NC_JUNC="yes"

### Minimum number of uniquely mapping reads needed to support junction (within or across samples)
UNIQUE_NUM=2


############################################
######    Read_Mapping  and QUANT    #######
############################################

#####IMPORTANT:
# Make sure that the "GEN_DIR" variable is supplied above (under Genome Index)
# This directory will also be used for read mapping

#	What are our QSub settings for Read_Mapping?
#	Below are the recommended settings
#	Note that STAR needs at least 30gb of memory to run
RM_QSUB="mem=50gb,nodes=1:ppn=16,walltime=450:00:00"

#   QUEUE=Slurm: What are our sbatch settings?
#       Below are the recommended settings
RM_PARTITION="batch"
RM_SBATCH="--nodes=1 --ntasks=16 --mem=50gb --time=45:00:00 --partition=${PARTITION}"


#   Where are the trimmed files? This handler will accept EITHER a text file list of forward samples to map (include the full file path for each sample) OR
#	a directory of input files (include the full file path). This may be the same as your output directory defined in adapter trimming (AT_OUTPUTDIR)
RM_INPUT="/scratch/ely67071/lettuce_dev_data/raw_rna_seq"
#       File name/path of where JOB IDs should be recorded
RM_JOB_LOG="/scratch/ely67071/lettuce_dev_data/RM_job_log_2nd_pass.txt"

#   What shared suffix do the forward samples have?
#       Example: _R1_paired.fq.gz
FORWARD="_1.fq.gz"

#   What shared suffix do the reverse samples have? (leave blank if single-end)
#       Example: _R2_paired.fq.gz
REVERSE="_2.fq.gz"

#   Where do you want your output files to go? Include the full file path
RM_OUTPUTDIR="/scratch/ely67071/lettuce_dev_data/STAR_output_gtf"

#	Define the number of threads to be used. 
#	This must be set to the number of available cores on the server node
RM_NTHREAD=16

#	What is the maximum number of mismatches allowed?
#	Alignment will output only if it has no more mismatches than this value
MAX_MIS=10

#	What is the maximum number of loci the read is allowed to map to?
#	All alignments will output only if the read maps to no more than this value
#	If greater than this value, read counted as "mapped to too many loci"
MAX_N=10

#	Do you want unmapped or partially mapped reads to output in separate fasta/fastq files?
#	If not, put "None", if yes, "Fastx"
UNMAP_F=Fastx

#	What is the minimum score needed for alignment?
#	Normalized by read length- (sum of mates' lengths for paired-end reads)
MINSCORE_READL=0.66

#	What is the minimum number of matched bases needed for alignment?
#	Normalized by read length- (sum of mates' lengths for paired-end reads)
MINMATCH_READL=0.66

#	What is the maximum length of read to use for seed search start points?
#	Reducing this parameter will increase the overall sensitivity of mapping.
SEEDSEARCH=50

#####	For adding read group (@RG) tags to mapped sequence:
#	This information is useful for differentiating reads coming from different runs/lanes even after merging

#	Name of Flowcell (or other name that differentiates a particular run, e.g. "Run1")
FLOWCELL_NAME="Run4"

# sequencing platform used
PLATFORM="ILLUMINA"

### Variables not shared with Collect_Junctions handler:

#	STAR outputs alignment in genome coordinates as well as transcript coordinates. The latter is used for
#	RSEM (step 7). If you plan on using the genomic coordinate alignment files for SNP calling, you will need 
#	sorted BAM files. If you want STAR to output genome alignments as sorted BAM files, put "yes" here. Note that this
#	increases the time it takes for read mapping to run. If "no", the genome alignment file will be an unsorted SAM.
GENOMIC_COORDINATE_BAMSORTED="yes"

#	The type of quantification requested
#	Either "TranscriptomeSAM" -outputs SAM/BAM alignments to transcriptome in a separate file
#	Or "GeneCounts" - read counts per gene
#	Can also put "-" for none
QUANT="GeneCounts"

#	2nd-pass read mapping will use junctions discovered in the first pass (Collect_Junctions)
#	This variable can be one of two options:
#		1.) The filtered junction list output from Collect_Junctions handler, OR
#		2.) If you have more than one filtered junction list (if Collect_Junctions was run with multiple sample groups),
#			this variable can be a text file listing the full file paths for each junction file you want included	
#	If not using additional junctions, leave blank
FILTERED_JUNC_LIST="/scratch/ely67071/lettuce_dev_data/STAR_output_gtf/junction_data/lettuce_dev__SJ.filtered.tab"

############################################
##########      Dependencies      ##########
############################################