##########################################################################################################
#RADIS version 1.0.0 (last update january 2016)                                                          #
#tested on Linux platforms (Ubuntu 15.10)                                                                #
#requirements                                                                                            #
# R (last version tested version 3.2.2 (2015-08-14) -- "Fire Safety")                                    #
# Stacks (last version tested 1.32, http://catchenlab.life.illinois.edu/stacks/source/stacks-1.32.tar.gz)#
# RAxML (last version tested 8.2.4, https://github.com/stamatak/standard-RAxML)                          #
#########################################################################################################

#-----------------------------------------------STEP 1.--------------------------------------------------# 
Download and unzip the tar.gz file
tar -xvzf RADIS-1.0.0.tar.gz
#--------------------------------------------------------------------------------------------------------#

#-----------------------------------------------STEP 2.--------------------------------------------------# 
Create a directory to store your raw illumina data and the barcodes_lib_names.txt file which provides
the list of barcodes (or combination of barcodes) you used to tag your samples and the corresponding 
sample codes (see examples in RADIS-1.0.0/datadir_example_single_barcode/barcodes_lib_names.txt or
RADIS-1.0.0/datadir_example_double_barcode/barcodes_lib_names.txt ; field separator = tab).
Please ...

• Note that sample codes MUST be different from each other to avoid troubles during renaming

• REMOVE empty lines in the barcodes_lib_names.txt file (remove blank lines at the end of files)

• Note that if you used Windows or MAC to create your barcodes_lib_names.txt files you may have trouble with the end of line characters.

This can generate an error after the filtering step (Cf the FAQ). Please use a DOS/MAC to Unix converter (dos2unix) to avoid such issues

mac2unix barcodes_lib_names.txt # will change format from Mac native to full UNIX
dos2unix barcodes_lib_names.txt # will change format from Windows to UNIX 
sed -i '/^$/d' barcodes_lib_names.txt # will remove extra spaces introduced by Mac 

#check your file before running :

wc -l barcodes_lib_names.txt # will count samples, check if result corresponds to your expectations
cat -A barcodes_lib_names.txt # (Linux) file, including field separators, is printed on screen
              # you should see ^I (tab) between columns and a $ at the end of each line and nothing else
              # more specifically you should not see any ^M
cat -te barcodes_lib_names.txt # (Mac) file, including field separators, is printed on screen
              # you should see ^I (tab) between columns and a $ at the end of each line and nothing else
              # more specifically you should not see any ^M
#--------------------------------------------------------------------------------------------------------#


#-----------------------------------------------STEP 3.--------------------------------------------------#
Fill up a RADIS.cfg configuration file with the parameters you chose to analyse your data. 
Examples .cfg files stored in RADIS-1.0.0/cfg may be used as templates 
DO NOT remove the # from the .cfg file
#--------------------------------------------------------------------------------------------------------#


#-----------------------------------------------STEP 4.--------------------------------------------------#
Create an output directory 
#--------------------------------------------------------------------------------------------------------#


#-----------------------------------------------STEP 5.--------------------------------------------------#
Run RADIS

# Data cleaning (filtering, demultiplexing and trimming of data. Removal of PCR duplicates and renaming of files)
(./bin/RADIS_step1_data_cleaning.pl --datadir rooted_directory_name --outdir rooted_directory_name [--cfg filename] [--ncpus integer] [--interleave_list_of_files] > outputfilename) 2> errorfilename 
 
# Data analysis (building individual loci, building catalog loci, selection of loci and samples and building phylip files for phylogenetic analysis. 
Phylip files are then analysed using RAxML. Workflow can be stopped before RAxML analyses using the flag --without-raxml)
(./bin/RADIS_step2_data_analyses.pl --listoffastqfiles 'quoted string' --outdir rooted_dirname [--cfg filename] [--ncpus integer] [--without-raxml] > outputfilename) 2> errorfilename 
 
# Whole pipeline (data cleaning + data analysis. Workflow can be stopped before RAxML analyses using the flag --without-raxml)
(./bin/RADIS.pl --datadir rooted_directory_name --outdir rooted_directory_name [--cfg filename] [--ncpus integer] [--interleave_list_of_files] [--without-raxml] > outputfilename) 2> errorfilename 
 
 
with:
 
--cfg : full path to the  configuration file

--datadir : full path to the input directory which should contain sequence files (.fq, .fq.gz, fastq, fastq.gz) and a barcode/sample code correspondance file

--interleave_list_of_files : use this flag when the ls command on the datadir directory groups all read1 files and then all read2 files (expected output from ls is file1_read1, file1_read2, file2_read1, file2_read2, etc)

--listoffastqfiles 'quoted string' : list of purified reads obtained after data cleaning. A text file listing absolute path to chosen fq_1 files can be provided (.fof). 

--ncpus : integer, number of cpus

--outdir : full path to the output directory.

--without-raxml : use this flag to stop the pipeline before RAxML analyses (just after .phy files)

outputfilename : will contain a report on the execution of all tasks

errorfilename : will list warning message and execution error
#----------------------------------------------------------------------------------------------------------------------------------#