#!/bin/bash
#SBATCH -t 4:00:00
#SBATCH --nodes=1 --ntasks-per-node=1
#SBATCH --mem=24g
#SBATCH --export=NONE
#SBATCH --mail-user=mramsey@uri.edu
#SBATCH --mail-type=BEGIN,END,FAIL

module load QIIME2/2019.7

tablein=/data/mramseylab/raw_reads/2018_Serum/Merge_runs/denoise-table-merge.qza
clsdir=/data/mramseylab/classifiers/
metadir=/data/mramseylab/metadata/
visdir=/data/mramseylab/visualizations/
filtdir=/data/mramseylab/proc_reads/
# filter status of input files, "MockQC" is just for the reads from the mock libraries
fil=MockQC


#Now to input the expected and actual read % belonging to control species in the samples
# expected values were made using Zymo data on composition.
# actual values were made by calculating based on the *.csv generated from the bar plot in the 1st script

biom convert \
  -i $filtdir$fil/Zymo_expected.tsv \
  -o $filtdir$fil/expected-taxonomy.biom \
  --table-type="OTU table" \
  --to-hdf5

qiime tools import \
 --type FeatureTable[RelativeFrequency] \
 --input-path $filtdir$fil/expected-taxonomy.biom \
 --output-path $filtdir$fil/expected-taxonomy.qza

biom convert \
  -i $filtdir$fil/Zymo_actual.tsv \
  -o $filtdir$fil/actual-taxonomy.biom \
  --table-type="OTU table" \
  --to-hdf5

qiime tools import \
 --type FeatureTable[RelativeFrequency] \
 --input-path $filtdir$fil/actual-taxonomy.biom \
 --output-path $filtdir$fil/actual-taxonomy.qza

qiime quality-control evaluate-composition \
  --i-expected-features $filtdir$fil/expected-taxonomy.qza \
  --i-observed-features $filtdir$fil/actual-taxonomy.qza \
  --o-visualization $filtdir$fil/$fil-6-comparison.qzv   --p-depth 6