Functional Annotation¶

Attach a Volume¶

Lets add a volume - as a reminder, this is sort of like plugging in an external harddrive to your laptop. Its adding a set of data, in this case, the databases that we are going to use for comparison for functional annotation.

This requires a few new steps in our tried and true Amazon EC2 instance protocol.

Choose AMI - Still going to choose Ubuntu server 14.04. Click Select.
Choose c4.2xlarge - Click Next:Configure Instance.
New Step For Subnet, select the one that ends in “us-east-1c”. Click Next: Add Storage.
Change Size 8Gb to 100Gb.
New Step Click add Volume.
- Enter snapshot ‘snap-d2d00056’
- Change to 100Gb
- And change to /dev/sdf
Click review and launch.

Now you can SSH into your instance as normal.

Make your Volume available.¶

We made the mount point /dev/xvdf.

udo mkdir /mnt/dbs
sudo mount /dev/xvdf /mnt/dbs/

Check out all the files you now have

ls /mnt/dbs/

Installs¶

Now start the install process. First install linux brew, as described here.

The rest of the installs will take ~10 minutes. Start by getting all the dependencies available in brew

brew install last blast emboss
brew install busco
brew install hmmer infernal

Next, there is a reciprocal best hit blast software that can be installed with the ruby gems package management system:

sudo gem install crb-blast

Transdecoder from source:

curl -LO https://github.com/TransDecoder/TransDecoder/archive/2.0.1.tar.gz
tar -xvzf 2.0.1.tar.gz
cd TransDecoder-2.0.1; make
export PATH=$HOME/TransDecoder-2.0.1:$PATH
cd

Now we’ll use pip, the python package management system to get the last few dependencies and dammit:

pip install -U setuptools
pip install numpy
pip install dammit
ln -s /home/linuxbrew/.linuxbrew/bin/busco /home/linuxbrew/.linuxbrew/bin/BUSCO_v1.1b1.py

Whew. Lets make sure it all worked.:

dammit dependencies

If all is well, it will say “All dependencies satisfied!”

And InterProScan (the second part of the lesson) needs java:

sudo add-apt-repository ppa:webupd8team/java
sudo apt-get update
sudo apt-get install oracle-java8-installer

This will give you a weird pink thing, where you hit enter for OK, left arrow to select Yes, then enter again.

Now set paths:

export JAVA_HOME=/usr/lib/jvm/java-8-oracle
export PATH=$PATH:/usr/lib/jvm/java-8-oracle

Dammit¶

Dammit is now installed but it doesn’t have any databases to compare anything to. These are also stored on the volume. Next step, tell dammit how to use the databases on the volume:

dammit databases --database-dir /mnt/dbs/dammit_dbs/ --full --busco-group eukaryota

Get example data and unzip:

wget ftp://ftp.ebi.ac.uk/pub/databases/pombase/FASTA/cdna_nointrons_utrs.fa.gz
gunzip cdna_nointrons_utrs.fa
grep -c '^>' cdna_nointrons_utrs.fa

Dammit takes quite a while on the whole set, so lets extract a smaller set to practice with:

head -1444 cdna_nointrons_utrs.fa > practice.fa
grep -c '^>' practice.fa

Run annotation (benchmark: 45 minutes for 50 genes with 8 threads):

dammit annotate practice.fa --database-dir /mnt/dbs/dammit_dbs/ --busco-group eukaryota --n_threads 15

You will get results from 5 different analyses:

Pfam-A
Rfam
OrthoDB
BUSCO
Transdecoder

To run with uniref90:

dammit annotate practice.fa --database-dir /mnt/dbs/dammit_dbs/ --busco-group eukaryota --n_threads 15 --full

InterProScan¶

Dammit runs a lot of good software, but you may want to also assign GO terms. InterProScan is one way to do this. It takes a while to install, so I put a copy on the volume. Lets add it to our path:

export PATH=$PATH:/mnt/dbs/interproscan-5.19-58.0

And see the documentation:

interproscan.sh | less

And we can make the software faster. It does not accept a parameter on the command line to increase the number of processors used, but it does have a properties file. Lets edit it.:

nano /mnt/dbs/interproscan-5.19-58.0/interproscan.properties

Change:

number.of.embedded.workers=6
maxnumber.of.embedded.workers=8

To:

number.of.embedded.workers=14
maxnumber.of.embedded.workers=15

Save with (Control-O, enter to save, Control-X to exit). It works with amino acid sequences, which we have from transdecoder (via dammit). (./practice.fa.dammit/practice.fa.transdecoder.pep). However, this file has astericks to represent stop codons, which IPS doesn’t like. We’ll make a copy and use our new sed skills to fix.:

cp ./practice.fa.dammit/practice.fa.transdecoder.pep practice.fa.transdecoder.nostars.pep
sed -i 's/\*//g' practice.fa.transdecoder.nostars.pep

And we will now make a results directory and run the software.:

mkdir ips_results
interproscan.sh \
-d ./ips_results \
-goterms \
-i ./practice.fa.transdecoder.nostars.pep \
-iprlookup \
-pa

Parameter breakdown:

Parameters

`-goterms`	lookup the GO terms associated with the database entry
`-iprlookup`	lookup the global InterPro accession number
`-pa`	lookup pathway annotation

Check out results

cd ips_results/
less practice.fa.transdecoder.nostars.pep.tsv

Notes on installing Interproscan¶

If the volume is no longer available and you need to install IPS, then here’s how to do it. Java needs to be installed (see above), then download and unpack the most recent version.:

wget ftp://ftp.ebi.ac.uk/pub/software/unix/iprscan/5/5.19-58.0/interproscan-5.19-58.0-64-bit.tar.gz
tar xvzf interproscan-5.19-58.0-64-bit.tar.gz

LICENSE: This documentation and all textual/graphic site content is licensed under the Creative Commons - 0 License (CC0) -- fork @ github. Presentations (PPT/PDF) and PDFs are the property of their respective owners and are under the terms indicated within the presentation.