#!/bin/bash
# make_HDF_dataset.sh - Convert tabular matrix to HDF5
# Extracted from 1k1RG Database Training Workshop 2025

indir=${1? Input directory with files pos.txt sample_list.txt and mat_vcf.txt}
proj=${2? Output (project) prefix, e.g. ricerp }

#loadmatrix=./bin/loadmatrix_geno.v2.hdf1.10    
loadmatrix=./loadmatrix_geno

n_snp=`wc -l $indir/pos.txt| awk '{print $1}' `
n_sampl=`wc -l $indir/sample_list.txt| awk '{print $1}' `

m=1000
n=64

if [[ ! -f "Matrix.txt" ]] ; then
ln -s $indir/mat_vcf.txt Matrix.txt
fi

# Transposed version: r=n_snp
r=$n_snp
$loadmatrix  -m $m -n $n -r $r \
-t \
 -o ${proj}_transp.h5 \
# -i $indir/mat_vcf.txt
