!wget http://ftp-trace.ncbi.nlm.nih.gov/sra/sdk/2.4.5-2/sratoolkit.2.4.5-2-ubuntu64.tar.gz

!tar -xvf sratoolkit.2.4.5-2-ubuntu64.tar.gz

!ls

!sratoolkit.2.4.5-2-ubuntu64/bin/fastq-dump SRR172903

!head -n 4 SRR172903.fastq

!wget https://github.com/agordon/fastx_toolkit/releases/download/0.0.14/fastx_toolkit-0.0.14.tar.bz2
!wget https://github.com/agordon/libgtextutils/releases/download/0.7/libgtextutils-0.7.tar.gz

!tar -xvf fastx_toolkit-0.0.14.tar.bz2
!tar -xvf libgtextutils-0.7.tar.gz

!bash fastx_install.sh

!fastq_quality_filter -q 33 -p 50 -i SRR172903.fastq > SRR172903.qc.fastq

!ls

!bash khmer-install.sh

!python unique-kmers.py -R unique_count -k 17 SRR172903.qc.fastq
!python unique-kmers.py -R unique_count -k 21 SRR172903.qc.fastq
!python unique-kmers.py -R unique_count -k 25 SRR172903.qc.fastq
!python unique-kmers.py -R unique_count -k 29 SRR172903.qc.fastq
!python unique-kmers.py -R unique_count -k 33 SRR172903.qc.fastq
!python unique-kmers.py -R unique_count -k 37 SRR172903.qc.fastq

!cat unique_count

!cat ncbi_acc.txt

!python fetch-genomes-fasta.py ncbi_acc.txt genomes

!wget http://sourceforge.net/projects/bowtie-bio/files/bowtie2/2.2.5/bowtie2-2.2.5-linux-x86_64.zip
!unzip bowtie2-2.2.5-linux-x86_64.zip

!apt-get install samtools

!bash bowtie.sh genomes/NC_000913.2.fa SRR172903.qc.fastq

!cat reads-mapped.count.txt

!cat reads-unmapped.count.txt

!cat reads.by.contigs.txt

!bash install-megahit.sh

!megahit/megahit --memory 10e9 -l 250 --k-max 81 -r SRR172903.qc.fastq --cpu-only -o megahit_assembly

!python khmer/sandbox/assemstats3.py 200 megahit_assembly/final.contigs.fa

!bash bowtie.sh megahit_assembly/final.contigs.fa SRR172903.qc.fastq

!cat reads-mapped.count.txt

!cat reads-unmapped.count.txt

!cat reads.by.contigs.txt

!wget ftp://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/2.2.30/ncbi-blast-2.2.30+-x64-linux.tar.gz
!tar -xvf ncbi-blast-2.2.30+-x64-linux.tar.gz

!cat genomes/*fa >> all-genomes.fa
!ncbi-blast-2.2.30+/bin/makeblastdb -in all-genomes.fa -dbtype nucl -out all-genomes
!ncbi-blast-2.2.30+/bin/blastn -db all-genomes -query megahit_assembly/final.contigs.fa -outfmt 6 -out contigs.x.all-genomes.blastnout

!head -n 10 contigs.x.all-genomes.blastnout

!wget http://downloads.sourceforge.net/project/fraggenescan/FragGeneScan1.19.tar.gz
!tar -xvf FragGeneScan1.19.tar.gz

!bash fraggenescan-install.sh

!FragGeneScan1.19/FragGeneScan -s megahit_assembly/final.contigs.fa -o final.contigs.orfs.fa -w 1 -t complete