scripts/Pcitri_tr2_rnaSpades_commands.txt

Version 1

#P. citri all data (VF+MF susphire and Edinburgh all Pcitri samples) RNAspades v3.13.0 transcriptome assembly
#run on bufo

cd /DATA/markop/
mkdir Pcitri_rnaspades_alldata
cd /DATA/markop/Pcitri_rnaspades_alldata


#unzip fastq files
cp /NFSs-datarepo/ngs_omics/NGS_pci_rna-seq_2018_mpe_Pcitri/data/*.gz ./input
cp /NFSs-datarepo/ngs_omics/NGS_pci_rna-seq_2019_mpe_SUSPHIRE-Edinburgh/data/*.gz ./input
pigz -d -v ./input/*.gz


#combine all SUSPHIRE (150 nt) and Edinburgh RNAseq reads (75 nt, 50 nt separately) into two files
cat ./input/MF_*_1.fastq ./input/VF_*_1.fastq > ./input/SUSPHIRE_1.fastq
cat ./input/MF_*_2.fastq ./input/VF_*_2.fastq > ./input/SUSPHIRE_2.fastq

cat ./input/PC_ALL_1_1.fastq ./input/PC_F_1_1.fastq ./input/PC_M_1_1.fastq > ./input/Edinburgh75_1.fastq
cat ./input/PC_ALL_1_2.fastq ./input/PC_F_1_2.fastq ./input/PC_M_1_2.fastq > ./input/Edinburgh75_2.fastq

cat ./input/PC_F_2_1.fastq ./input/PC_M_3_1.fastq > ./input/Edinburgh50_1.fastq
cat ./input/PC_F_2_2.fastq ./input/PC_M_3_2.fastq > ./input/Edinburgh50_2.fastq

rm ./input/MF*.fastq
rm ./input/VF*.fastq
rm ./input/PF*.fastq
rm ./input/PC*.fastq


#pipeline for cleaning reads suggested here: http://seqanswers.com/forums/showthread.php?p=204724#post204724

#Trim adapters
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/SUSPHIRE_1.fastq in2=./input/SUSPHIRE_2.fastq out=./input/SUSPHIRE_trimmed.fastq ktrim=r k=23 mink=11 hdist=1 ref=/DATA/markop/bbmap/resources/adapters.fa tbo tpe maxns=0 trimq=20 qtrim=r maq=12
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/Edinburgh50_1.fastq in2=./input/Edinburgh50_2.fastq out=./input/Edinburgh50_trimmed.fastq ktrim=r k=23 mink=11 hdist=1 ref=/DATA/markop/bbmap/resources/adapters.fa tbo tpe maxns=0 trimq=20 qtrim=r maq=12
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/Edinburgh75_1.fastq in2=./input/Edinburgh75_2.fastq out=./input/Edinburgh75_trimmed.fastq ktrim=r k=23 mink=11 hdist=1 ref=/DATA/markop/bbmap/resources/adapters.fa tbo tpe maxns=0 trimq=20 qtrim=r maq=12

#Remove small contaminants
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/SUSPHIRE_trimmed.fastq out=./input/SUSPHIRE_filtered.fastq k=31 ref=/DATA/markop/bbmap/resources/sequencing_artifacts.fa.gz,/DATA/markop/bbmap/resources/phix174_ill.ref.fa.gz
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/Edinburgh50_trimmed.fastq out=./input/Edinburgh50_filtered.fastq k=31 ref=/DATA/markop/bbmap/resources/sequencing_artifacts.fa.gz,/DATA/markop/bbmap/resources/phix174_ill.ref.fa.gz
/DATA/markop/bbmap/bbduk.sh -Xmx230g in=./input/Edinburgh75_trimmed.fastq out=./input/Edinburgh75_filtered.fastq k=31 ref=/DATA/markop/bbmap/resources/sequencing_artifacts.fa.gz,/DATA/markop/bbmap/resources/phix174_ill.ref.fa.gz

rm ./input/*_trimmed.fastq

# SPAdes BayesHammer error correction
#  you cannot specify --only-error-correction in RNA-Seq mode!

/home/administrator/Software/SPAdes-3.13.1-Linux/bin/spades.py \
--only-error-correction \
-m 230 \
-t 32 \
-o ./input \
--pe1-fr --pe2-fr --pe3-fr \
--pe1-12 ./input/SUSPHIRE_filtered.fastq \
--pe2-12 ./input/Edinburgh50_filtered.fastq \
--pe3-12 ./input/Edinburgh75_filtered.fastq

rm -rf ./input/tmp
rm -rf ./input/split_input

#Error-correct 1
/DATA/markop/bbmap/bbmerge.sh -Xmx230g in1=./input/corrected/SUSPHIRE_filtered_1.00.0_0.cor.fastq.gz in2=./input/corrected/SUSPHIRE_filtered_2.00.0_0.cor.fastq.gz out=./input/SUSPHIRE_ecco_PE.fastq ecco mix vstrict adapters=default
/DATA/markop/bbmap/bbmerge.sh -Xmx230g in=./input/corrected/SUSPHIRE_filtered__unpaired.00.0_0.cor.fastq.gz out=./input/SUSPHIRE_ecco_SE.fastq ecco mix vstrict adapters=default

/DATA/markop/bbmap/bbmerge.sh -Xmx230g in1=./input/corrected/Edinburgh50_filtered_1.00.1_0.cor.fastq.gz in2=./input/corrected/Edinburgh50_filtered_2.00.1_0.cor.fastq.gz out=./input/Edinburgh50_ecco_PE.fastq ecco mix vstrict adapters=default
/DATA/markop/bbmap/bbmerge.sh -Xmx230g in=./input/corrected/Edinburgh50_filtered__unpaired.00.1_0.cor.fastq.gz out=./input/Edinburgh50_ecco_SE.fastq ecco mix vstrict adapters=default

/DATA/markop/bbmap/bbmerge.sh -Xmx230g in1=./input/corrected/Edinburgh75_filtered_1.00.2_0.cor.fastq.gz in2=./input/corrected/Edinburgh75_filtered_2.00.2_0.cor.fastq.gz out=./input/Edinburgh75_ecco_PE.fastq ecco mix vstrict adapters=default
/DATA/markop/bbmap/bbmerge.sh -Xmx230g in=./input/corrected/Edinburgh75_filtered__unpaired.00.2_0.cor.fastq.gz out=./input/Edinburgh75_ecco_SE.fastq ecco mix vstrict adapters=default

rm ./input/*_filtered.fastq



#Error-correct 2
#used parameters that were suggested by Brian Bushnell here: http://seqanswers.com/forums/archive/index.php/t-72901.html and https://www.biostars.org/p/225338/
/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/SUSPHIRE_ecco_PE.fastq out=./input/SUSPHIRE_eccc_PE.fastq ecc passes=6 minid=0.98
/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/SUSPHIRE_ecco_SE.fastq out=./input/SUSPHIRE_eccc_SE.fastq ecc passes=6 minid=0.98

/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/Edinburgh50_ecco_PE.fastq out=./input/Edinburgh50_eccc_PE.fastq ecc passes=6 minid=0.98
/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/Edinburgh50_ecco_SE.fastq out=./input/Edinburgh50_eccc_SE.fastq ecc passes=6 minid=0.98

/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/Edinburgh75_ecco_PE.fastq out=./input/Edinburgh75_eccc_PE.fastq ecc passes=6 minid=0.98
/DATA/markop/bbmap/clumpify.sh -Xmx200g in=./input/Edinburgh75_ecco_SE.fastq out=./input/Edinburgh75_eccc_SE.fastq ecc passes=6 minid=0.98

#Error-correct 3
/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/SUSPHIRE_eccc_PE.fastq out=./input/SUSPHIRE_ecct_PE.fastq ecc
/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/SUSPHIRE_eccc_SE.fastq out=./input/SUSPHIRE_ecct_SE.fastq ecc

/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/Edinburgh50_eccc_PE.fastq out=./input/Edinburgh50_ecct_PE.fastq ecc
/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/Edinburgh50_eccc_SE.fastq out=./input/Edinburgh50_ecct_SE.fastq ecc

/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/Edinburgh75_eccc_PE.fastq out=./input/Edinburgh75_ecct_PE.fastq ecc
/DATA/markop/bbmap/tadpole.sh -Xmx200g in=./input/Edinburgh75_eccc_SE.fastq out=./input/Edinburgh75_ecct_SE.fastq ecc

rm ./input/*_ecco_*.fastq
rm ./input/*_eccc_*.fastq

#Merge
/DATA/markop/bbmap/bbmerge.sh -Xmx200g in=./input/SUSPHIRE_ecct_PE.fastq out=./input/SUSPHIRE_merged.fastq outu=./input/SUSPHIRE_unmerged.fastq rem k=62 extend2=50 adapters=default
/DATA/markop/bbmap/bbmerge.sh -Xmx200g in=./input/Edinburgh50_ecct_PE.fastq out=./input/Edinburgh50_merged.fastq outu=./input/Edinburgh50_unmerged.fastq rem k=62 extend2=50 adapters=default
/DATA/markop/bbmap/bbmerge.sh -Xmx200g in=./input/Edinburgh75_ecct_PE.fastq out=./input/Edinburgh75_merged.fastq outu=./input/Edinburgh75_unmerged.fastq rem k=62 extend2=50 adapters=default


#Split unmerged into two files
# not needed /DATA/markop/bbmap/reformat.sh in=./input/SUSPHIRE_unmerged.fastq out1=./input/SUSPHIRE_unmerged_1.fastq out2=./input/SUSPHIRE_unmerged_2.fastq overwrite=true

rm ./input/corrected/*.fastq.gz
rm ./input/*._ecct_PE.fastq

#running rnaSPAdes (BayesHammer error correction turned off)
# rnaSPAdes can take as an input only paired-end and single-end libraries; merged will therefore be treated as single!
# rnaSPAdes does not support --careful and --cov-cutoff options.
# By default rnaSPAdes uses 2 k-mer sizes, which are automatically detected using read length (approximately one third and half of the maximal read length). We recommend not to change this parameter because smaller k-mer sizes typically result in multiple chimeric (misassembled) transcripts.

ulimit -n 32000

cat ./input/SUSPHIRE_merged.fastq ./input/SUSPHIRE_ecct_SE.fastq > ./input/SUSPHIRE_SE.fastq
cat ./input/Edinburgh50_merged.fastq ./input/Edinburgh50_ecct_SE.fastq > ./input/Edinburgh50_SE.fastq
cat ./input/Edinburgh75_merged.fastq ./input/Edinburgh75_ecct_SE.fastq > ./input/Edinburgh75_SE.fastq

/home/administrator/Software/SPAdes-3.13.1-Linux/bin/rnaspades.py \
--only-assembler \
-k 29,49 \
-m 245 \
-t 32 \
-o ./output \
--ss-rf \
--pe1-fr --pe2-fr --pe3-fr \
--pe1-12 ./input/SUSPHIRE_unmerged.fastq \
--pe1-s ./input/SUSPHIRE_SE.fastq \
--pe2-12 ./input/Edinburgh50_unmerged.fastq \
--pe2-s ./input/Edinburgh50_SE.fastq \
--pe3-12 ./input/Edinburgh75_unmerged.fastq \
--pe3-s ./input/Edinburgh75_SE.fastq

rm ./input/*.fastq

#produced xx hard-filtered, x normal-filtered, and x soft-filtered transcripts.
#should be mapped to genome scaffolds

#Evaluate assemlby
/DATA/markop/bbmap/stats.sh -Xmx200g in=./output/soft_filtered_transcripts.fasta out=./output/soft_filtered_transcripts.bbmap.stats