jelber2

# Revisions 1 & 2 use the same read name read1 and read2, this
# revision 3 appends _1 and _2 to read1 and read2, respectively.
#
# Basic idea is to take a long read alignment file that, then shred the long reads into shreds, keeping their alignment information
# output hi-c-like read-pairs, interleaved in a BAM file!
#
# So for example,
# ----------------------- 1long read
# 1r1 2r1 3r1  3r2 2r2 1r2
# results in        1read1, 1read2

jelber2

# revision 2, now supports multi-core processing
# Basic idea is to take a long read alignment file that has been shredded with shredBAM.jl, then
# output hi-c-like read-pairs
#
# So for example,
# ----------------------- 1long read
# 1r1 2r1 3r1  3r2 2r2 1r2
# results in        1read1, 1read2
#                   2read1, 2read2
#                   3read1, 3read2

jelber2

# Get HG002
# wget https://s3-us-west-2.amazonaws.com/human-pangenomics/T2T/HG002/assemblies/hg002v1.0.1.fasta.gz
# gunzip hg002v1.0.1.fasta.gz
#
# Use seqtk to get maternal and paternal sequences
# https://github.com/lh3/seqtk
# seqtk seq -l0 hg002v1.0.1.fasta|paste - - |fgrep "MATERNAL" |tr '\t' '\n'|seqtk seq -l60 > hg002v1.0.1.maternal.fasta &
# seqtk seq -l0 hg002v1.0.1.fasta|paste - - |fgrep "PATERNAL" |tr '\t' '\n'|seqtk seq -l60 > hg002v1.0.1.paternal.fasta &
#
# use samtools to index haplotypes

jelber2

# new in revision 2, multithreaded support - up to twice as fast as single-threaded, revision 1
# invoke with:

# julia --threads 16 Plot_Animated_Nanopore_Quality_x_Length_2D_Contours.jl --input_file test.bam --fps 0.7 --output_file test.gif

# but need a lot of threads (up to 16, above 16 threads, get saturation in cores vs. time plot) 

using XAM
using ArgParse
using DataFrames

jelber2

# note: the Dorado BAM file produced by basecalling POD5 files needs to be indexed with "samtools index input_file"
# note2: something seems really strange about the qualtiy scores, I have checked several times, but they seem correct
# I am not sure why they are reporting that for example reads with Q38 average quality scores.
# EDIT: this is fixed, see https://www.biostars.org/p/295932/#295936 for better formula for average Phred Score
# tested with julia-1.10.3 and XAM.jl-0.40
using XAM
using ArgParse
using DataFrames
using Dates
using Plots

jelber2

# note: outputs to STDOUT a SAM file without a header
# note2: remove auxillary tags and information
# note3: ignores RNEXT and PNEXT from BAM file (puts * and 0 respectivelv)
# tested with julialang v1.10.2 and XAM v0.4.0
# $ julia changeBAMquality.jl input.bam ? > test.sam.noheader
#
# to add a header from the original BAM file that had its qualities changed
# and add on a PG line
# $ ASCII_CHARACTER="?"
# $ INPUT=input.bam

jelber2

using XAM
using ArgParse
using DataFrames
using Dates
using Plots
using StatsBase
using Plots.PlotMeasures


function parse_commandline()

jelber2

# tested on Julialang v.1.10.2, DataStructures v0.18.16, and FASTX v2.1.4
# Usage
# $ julia stitch-fasta.jl chr20.herro.fasta.Q30.recal.shred.fasta > chr20.herro.fasta.Q30.recal.fasta
#
import Pkg; Pkg.add("FASTX")
import Pkg; Pkg.add("DataStructures")
using DataStructures
using FASTX

function process_fasta_file(filename::String)

jelber2

# tested on Julialang v.1.10.2 , DataStructures v0.18.16, and FASTX v2.1.4
# Usage
# $ julia stitch-fastq.jl chr20.herro.fasta.Q30.recal.shred.fastq > chr20.herro.fasta.Q30.recal.fastq
#
import Pkg; Pkg.add("FASTX")
import Pkg; Pkg.add("DataStructures")
using DataStructures
using FASTX

function process_fastq_file(filename::String)

jelber2

# previous versions allowed to generate directly from BAM, but there seemed to have been some troubles 
# between versions of XAM, so this version is working so far for its intended purpose on https://github.com/brendanofallon/jovian
# also, this version is rather fast
#
#
# note: ignores quality scores at the moment - fixed in revision#4
# note2: outputs to STDOUT a SAM file without a header
# tested with julialang v1.10.2 and XAM v0.4.0
# $ julia shredBAM.jl input.bam 300 > test.sam.noheader
#