BLASTing your assembled data

One thing everyone wants to do is BLAST sequence data, right? Here’s a simple way to set up a stylish little BLAST server that lets you search your newly assembled sequences.

Installing blastkit

Installing some prerequisites:

pip install pygr
pip install whoosh
pip install git+
pip install git+
apt-get -y install lighttpd

and configure them:

cd /etc/lighttpd/conf-enabled
ln -fs ../conf-available/10-cgi.conf ./
echo 'cgi.assign = ( ".cgi" => "" )' >> 10-cgi.conf
echo 'index-file.names += ( "index.cgi" ) ' >> 10-cgi.conf

/etc/init.d/lighttpd restart

Next, install BLAST:

cd /root

curl -O
tar xzf blast-2.2.24-x64-linux.tar.gz
cp blast-2.2.24/bin/* /usr/local/bin
cp -r blast-2.2.24/data /usr/local/blast-data

And put in blastkit:

cd /root
git clone -b ec2
cd blastkit/www
ln -fs $PWD /var/www/blastkit

mkdir files
chmod a+rxwt files
chmod +x /root

and run

cd /root/blastkit
python ./

It should say everything is OK.

Adding the data

Take an assembly of interest, and copy it into /root/blaskit/db/db.fa; e.g.

cp /mnt/assembly/ecoli.21/contigs.fa /root/blastkit/db/db.fa

Formatting the database

After you’ve done either of the above, format and install the database for blastkit:

cd /root/blastkit
formatdb -i db/db.fa -o T -p F
python db/db.fa



You can install any file of DNA sequences you want this way; just copy it into /root/blastkit/db/db.fa and run the indexing commands, above.

Running blastkit

Figure out what your machine name is (ec2-???-???-???-??? and go to:


Make sure you have enabled port 80 in your security settings on Amazon.

...and you should see a BLAST page. If you’re searching E. coli, try:

comments powered by Disqus