BRCA1 bioinformatic analysis

Bioinformatic project for the namesake subject at ITBA. We chose as our gene to analyze the BRCA1 gene highly related with breast cancer development.

Deliverables

We included in this repository a Report.pdf which includes result analyisis, development description and answers for the theoretic exercises (Such as 2b and 3 exercises). We also included a Presentation.pdf file, containing the slides to showcase our conclusions and our work in this project.

Requirements

• Ruby 2.5.1
• Bundler
• EMBOSS
• Prosite database

Set up

The project was developed in Ruby. We used Bundler to install the required gems to run the project. The only gem we use in the development was Bioruby as recommended by the professor. So, to install the project we only have to install the required gem.

bundle install --path ./vendor/bundle

ALERT: The last week there was a breaking change to the BLAST remote server (they started using https) and the BioRuby library is not working while doing remote queries to BLAST. Thus, Ex2.rb may not work if the issue was not fixed. Please follow the issue we opened on BioRuby's github repository to keep updated.

However we found a fix to the issue, but you will have to edit the code for the downloaded gem. After installing ruby gems, you will have to edit 2 files:

On vendor/bundle/ruby/2.5.0/gems/bio-1.5.1/lib/bio/appl/blast/genomenet.rb change the line 214 (plus means add, minus means remove the line):

- http = Bio::Command.new_http(host)`) for:
+ http = Bio::Command.new_https(host) 

On vendor/bundle/ruby/2.5.0/gems/bio-1.5.1/lib/bio/command.rb add this code in line 793:

 def new_https(address, port = 443)
    uri = URI.parse("https://#{address}")
    # Note: URI#find_proxy is an unofficial method defined in open-uri.rb.
    # If the spec of open-uri.rb would be changed, we should change below.
    if proxyuri = uri.find_proxy then
      raise 'Non-HTTP proxy' if proxyuri.class != URI::HTTP
      Net::HTTP.new(address, port, proxyuri.host, proxyuri.port)
    else
      d= Net::HTTP.new(address, port)
      d.use_ssl=true
      d
    end
  end

This is a hackish solution but is enough to make Ex2.rb run.

Excersise 1

We have to read a nucleotids sequence from a mRNA from the BRCA1 gene in GenBank format. Then, translate this sequence to the possible amino acids sequences (given the 6 possible reading frames) and finally writing the output to FASTA files.

The script Ex1.rb given a GenBank gene file (BRCA1.gb in this case) generates 6 files with the possible amino acid sequences with the names <mNRA_ID>$<FRAME_NUM>.fas.

bundle exec ruby Ex1.rb BRCA1.gb

Excersise 2

We have to run a BLAST for each one of the 6 possible amino acids sequences generated in the last exercise.

The script Ex2.rb given a mNRA id (NM_007294 in this case) generate 6 files with the results of the BLAST execution. It reads the files generated in the previous exercise and generate a file <mNRA_ID>$<FRAME_NUM>.blas with the report hits for each one. (It's slow)

bundle exec ruby Ex2.rb NM_007294

Excersise 4

We developed a script that given a pattern it searches for possible matches in the hits from a BLAST report.

The script Ex4.rb given a BLAST file (NM_007294$2.blas in this case), an output file, and a pattern, generates a report with the matching hits with the given pattern and its FASTA file.

bundle exec ruby Ex4.rb NM_007294.xml NM_007294$2.match SAPIENS --protein

Excersise 5

We developed a script that given the nucleotids sequence from a GenBank file, will return the possible amino acid sequences for each ORF.

The script Ex5.rb given a GenBank gene file (BRCA1.gb in this case) will generate an output file (input) with all the possible amino acid sequeneces for each ORF. Its inputs are the option --orf, the GenBank file and the output file.

bundle exec ruby Ex5.rb --orf BRCA1.gb BRCA1.orf

On the the other hand, the same script also generate a domain analyisis for the given amino acid sequences, in a FASTA file. Its inputs are --prosite, the FASTA file, and the output file. To run this script you will need the prosite database downloaded in the root folder within a folder named prosite.

bundle exec ruby Ex5.rb --prosite NM_007294\$1.fas analysis.prosite

Notes

The code is developed to be non-dependant of the analyzed gene, making it a useful generic tool.

To install EMBOSS download the tar file, unzip it and then run:

./configure --without-x
make
make install

Developers

• Ramiro Olivera Fedi [email protected]
• Julian Antonielli [email protected]