#!/bin/bash

## This is an example configuration script for Ginkgo. When using Ginkgo via the webserver, this is
## in charge of creating this file. This example is only useful if you are planing to run Ginkgo
## as a standalone programme.


# =======================================================================================
# Running options
# =======================================================================================
# init=1 -> Clean the directory and start from scratch the whole analysis
init=1
# process=1 -> Run mapped data through primary pipeline
process=1
# fix=1 -> Recreate clusters/heat maps (not required if process=1)
fix=0
# =======================================================================================


# =======================================================================================
# Segmentation
# =======================================================================================
# Use one of these methods to segment:
# 0 -> Independent (normalized read counts)
# 1 -> Global (sample with lowest IOD)
# 2 -> Custom (using uploaded reference sample)
segMeth=2
# =======================================================================================


# =======================================================================================
# Genome
# =======================================================================================
# Directory name for ROOT_DIR/genomes/${chosen_genome}
# Pre-calculated files are only available for hg19
chosen_genome=hg19
# =======================================================================================


# =======================================================================================
# Mask Y-chr pseudoautosomal regions (experimental)
# =======================================================================================
# if 1, use ROOT_DIR/genomes/${chosen_genome}/pseudoautosomal
# else -> use ROOT_DIR/genomes/${chosen_genome}/original
# Pre-calculated files are only available for "original", set it to "0"
rmpseudoautosomal=0
# =======================================================================================


# =======================================================================================
# Genome bins
# =======================================================================================
# This is a complex value made of the concatenation of
# - type: variable or fixed (bins. Variable refers to amount of mappable genome, recommended)
# - size: available values are 10000000, 5000000, 2500000, 1000000, 500000, 250000, 175000, 100000,
#       50000, 25000, 10000
# - read-length: available values are: 150, 101, 76, 48
# - aligner: bowtie or bwa
# The read-length and aligner refer to the simulations of re-mapping reads of that length with that
# aligner on the whole genome. This is used to calculate bins of "mappable" (i.e. variable) genome.
# The resulting value is the name of a file under ginkgo/genomes/$choosen_genome/original/ with the
# bin coordinates
binMeth=variable_1000000_150_bowtie

# Mask bad bins (experimental)
# Removes bins with consistent read pileups from the analysis (e.g. at chromosome boundaries)
rmbadbins=0

# Probably some legacy options. Always set to 0 and empty on the Ginkgo webserver. This seemed to be
# an option to provide user-defined bins.
b=0
binList=
# For user-defined segmentation (seems to be disabled now)
ref=XX_NG_ICM_E_4M.median.bed
# =======================================================================================


# =======================================================================================
# FACS file
# =======================================================================================
# Options to specify a user-provided FACS file (for ploidy)
# f=0 if not FACS file is provided. f=1 if FACS file is provided.
f=0

# User-provided FACS file with cell name and ploidy
facs=
# =======================================================================================


# =======================================================================================
# Clustering
# =======================================================================================
# Distance measure. Options can be:
# - euclidean
# - maximum
# - manhattan
# - canberra
# - binary
# - minkowsky
# This is the distance measure used to calculate the distance matrix with the dist function from the
# R package stats
distMeth=euclidean

# Clustering method. Options can be:
# - ward (best to either use ward.D or ward.D2 if using a modern version of R)
# - single
# - complete
# - average
# - NJ
# This is the method used to calculate the dendrogram with the hclust function from the R package
# stats, except for NJ for which the ape library is used.
clustMeth=ward.D2

# Include sex chromosome (1: yes; 0: no). Probably safe to leave it as 1 unless you have a mixture
# of male and female cells in which case it might be a good idea to leave sex chromosomes out.
sex=1
# =======================================================================================


# =======================================================================================
# Display options
# =======================================================================================
# Plot gene locations (1) or not (0)
q=0

# Color scheme:
# - 3: dark blue / red
# - 1: light blue / orange
# - 2: magenta / gold
color=3
# =======================================================================================


# =======================================================================================
# Reporting by email
# =======================================================================================
# Only required if you want to get an email at the end of the run. Probably not necessary
# in standalone mode. Set email to nothing to skip email feature. Otherwise, set your
# email address and the permalink which is meant to be the location of the results.
user=
email=
permalink=
# =======================================================================================