### D. virilis recombination msg file

# For non-cluster usage, set cluster to 0
cluster=1
queue=1day

# List of barcodes (columns = barcode/tplate_id/cross_id/tsex) - see posted example
barcodes=barcodes1.3

# Illumina index file parameters: (optional) 
# Should be a fastq/fasta formatted file (can be gz).
# index_file=idx.fastq
# A list of labels with barcodes (tab separated) for each index you want to parse out
# index_barcodes=indexes.txt

# Use an experimental (faster) parser based on grep (1 for yes, 0 for no)
# This should only be used if your barcode is at the beginning of the read and 
# you want to keep everything after the barcode.
new_parser=0
# If using the new parser (see above), this chops of this number of base pairs
# after the barcode.
new_parser_offset=0
# If using the new parser enter an optional sequence to filter out such as a dimer
# adaptor.
# new_parser_filter_out_seq=GTGCGAGGAGGCTTGAAGATCGGAA

# Barcoded reads from a genetic cross
reads=indivDys00_A01_ATGCATGC.fastq 

# Set to 1 for more verbose output messages, and to prevent MSG to from deleting intermediate/temp files
# Warning: Setting this to 1 will consume substantially more disk space.
debug=0

# Set to 1 to get plots of summarized result data: lod-matrix.bmp, missing.pdf, segregation.pdf,
# and offdiagonal-lod.pdf.  (runs summaryPlots.R module)
# Set to 0 to run a faster module (combine.py) and reduce memory usage.  It still produces
# ancestry-probs-par1.tsv, and ancestry-probs-par2.tsv files.
full_summary_plots=1

# Restriction enzymes recognized: MseI|NdeI|Hpy188I|HphI|MboII|XcmI|Hpy188III|AhdI|HpyAV|null
re_cutter=null #MseI first batch
# linker_system enzymes recognized: Tn5-IonTorrent|Dros_SR_vII
# (Set re_cutter above to null when using Tn5-IonTorrent here)
linker_system=Tn5-IonTorrent #Dros_SR_vII first batch

# Parental reference genomes
# Note that all chromosome/contig information and genome coordinates are taken from parent1 genome
# If you have updated the parental genomes previously, then use the names of the updated genomes here.
parent1=s9_2.fasta
parent2=s160_2.fasta

# Set max_mapped_reads to limit the number of reads.  This will keep the first <max_mapped_reads> lines
# in the sam files.  Benefits to using this could be to reduce computation time, the resoution of 
# breakpoints tends to asymptote after a certain number of reads, so adding more reads doesn’t 
# provide more information.  We also reduce the chance that one-read_per_contig ends up undersampling the genome.
# Individuals with the most reads often produce small, clearly incorrect, hmm prob shifts where there is a 
# pile of improper mapping. Perhaps repeetitive regions? Anyway, for individuals with fewer reads the hmm plows 
# right through these regions.
# NOTE: This option will truncate the sam files, so to try different # reads, you need to delete the *sam_files folder
max_mapped_reads=10000

# Algorithm for constructing BWT index: bwtsw (default) or is (for parental 
# genomes smaller than 10MB)
bwaindex1=bwtsw
bwaindex2=bwtsw

# Algorithm for BWA mapping (aln or bwasw)
# aln should be faster (and more suited for short reads).  bwasw may be more
# accurate. (This is ignored if use_stampy is set to 1)
bwa_alg = aln

# Number of threads used by BWA (multi-threading mode)
bwa_threads = 1

# Use stampy instead of BWA for mapping.  (1 or 0)
use_stampy = 0
# Pre-map with BWA before running stampy. This is almost always recommended. (1 or 0)
# (This is ignored if use_stampy = 0 since you wouldn't be using stampy to begin with..)
stampy_premap_w_bwa = 1

# If running stampy, optionally specify how many pseudo threads it will use
# this is only valid with the cluster option set to 1.
# set this to 0 to not use this. 
stampy_pseudo_threads = 0

#theta indicates the amount of dependence between separate reads. If all reads are completely
#independent, then theta= 1. If reads show some dependence, for example due to PCR duplication, then theta 
#should be < 1. Valid values are 0 < theta ≤ 1.
#See doi:10.1101/gr.9.9.868 and doi:10.1101/gr.078212.108 for further details.
theta=1

# Chromosomes to analyse, sex chromosomes (male hemizygous) and those 
# to produce plots for
# Note: If you genome contains many contigs, setting "chroms = all" may cause msg to take a long time to finish. 
# Also, it is not clear how reliable the msg results will be for smaller contigs.
# It is often expeditious to limit the analysis to only chromosomes and larger contigs.
# So, for Drosophila, perhaps set "chroms = 2L,2R,3L,3R,4,X"
chroms=all
sexchroms=A
chroms2plot=all

# Priors on ancestry of autosomes, in following order
# Homozygous parent1, heterozygous, homozygous parent2
# For sex-linked chromosomes in males, the priors become (0.5, 0, 0.5).
# F1 x F1 cross: 0.25,0.5,0.25 
# F1 x parent2 backcross: 0,0.5,0.5
priors=0.5,0.5,0

# These two parameters determine the value of the "allele frequencies" (P_{kmj} in the paper)==gamma
# Essentially they give the probability that the study strain does not have 
# the same allele as the (perhaps updated) reference sequence, for the two parental strains. 
# This parameter is estimated from putatively homozygous regions and a plot with the
# error estimates for both parents can be found in hmm_fit_images (default 0.1)
deltapar1=.01
deltapar2=.01

# MSG will output a GFF file (for importing into Geneious) for each individual
# for each chromosome.  This value specifies the confidence threshold to use 
# when selecting breakpoints to be included in the Geneious files.  The default is 
# .95 (95%) which means it selects breakpoints at 5% and 95% confidence.
gff_thresh_conf=.95

# Recombination rate
# Expected number of recombination events per genome per meiosis 
# If recRate is =0, then 1 recomb per contig is used.
# If recRate is specified != 0, then (recRate / genome) is used.
recRate= 5

# This parameter can be used to alter the recombination rate in the model:
# if it is greater than one then the model will be tend to introduce more recombination
# breakpoints; less than one and it will introduce fewer. (default 1)
rfac=.00001

# Minimum difference between posterior probs to thin (summaryPlots.R)
difffac=0.01
# Min fraction of missing data (summaryPlots.R)
pnathresh=.03

# For proportion emmision probs threshold.  (what percent of individuals should 
# cover a chromosome site to be included in hmm_fits_ests.csv output file.)
# Example .5 is 50%
# (Leave unspecified if you don't need this. It will save some disk space)
#pepthresh=.75

# Used in fitting HMM.  Use 0 for FALSE, 1 for True
one_site_per_contig=1

# Should all reads be quality trimmed and what values to use. 
# Set quality_trim_reads_thresh to 0 to skip this step.
# Base intensity threshold value (Phred quality scores 0 to 40

quality_trim_reads_thresh = 10

# Minimum number of consecutive bases passing threshold values

quality_trim_reads_consec = 30

#When using Stampy, set divergence for mapping reads to a reference (note: this is
#strongly recommended by stampy for divergences >3%, as it will automatically
#shut down BWA pre-mapping which otherwise cause occasional segfaults)
#(stampy default if .001) #comment out to use stampy default.
#This is in %. For example if you want 5% set these values to 0.05
indiv_stampy_substitution_rate = 0.05

# Filter out poor alignments, set this to 0 to skip
indiv_mapq_filter = 20

# Custom qsub options.  Modify these if you are running on a customized SGE with
# special options for requesting an exclusive node. The commented out options below
# are what are used for the cluster at Janelia Farm.
# These default to blank if they are not specified.
# for the pe option leave off the slot count, this is set from the threads option.
#addl_qsub_option_for_exclusive_node=-l mem96=true,excl=true
custom_qsub_options_for_all_cmds=-q sixhour -l nodes=2:ppn=10:ib,mem=20gb,walltime=6:00:00
#addl_qsub_option_for_pe=-pe batch
#threads=8

# Send an email alert when the MSG run is complete/succeeds (optional)
# Warning: an email will NOT be sent if there is an error and MSG does not
# finish.
# (Separate multiple email addresses with ;)
#email_host=10.11.5.81
#notify_emails=test@example.com