Gianluca Croso, Heather Han, Nitin Kumar, Eric Walker
EN.600.439 Computational Genomics
Final Project: README

How to run the code:
Encode (compress) for variant:  python3 jam.py e v samfile.sam refgenome.fa
Decode variant compressed file: python3 jam.py d v jamfile.vjam refgenome.fa
Encode (compress) for peak:     python3 jam.py e p samfile.sam refgenome.fa
Decode peak compressed file:    python3 jam.py d p jamfile.pjam refgenome.fa

To reproduce our results:
The ex1 files for fast testing (small sam file and reference genome),
as well as the simulation data and ecoli reference genome used are provided
with the sumbission. They are, respectively:
ex1.sam
ex1.fa
sim1.sam
sim1.fa
Ecoli1.fa

The first 4 files are for fast testing. The full E. Coli genome and the other
files linked below take a significantly longer time to run.

The drosophila genome is the dm6 reference genome available at
support.illumina.com/sequencing/sequencing_software/igenome.html

The Ecoli genome was also obtained at the same website, but we had
to alter the sequence title to match the sequence name in our sam
dataset, and that is why we provide the E. Coli genome here.

The Ecoli read-alignment file is available at:
ftp://webdata:[email protected]/Data/SequencingRuns/DH10B/MiSeq_Ecoli_DH10B_110721_PF.bam

The Drosophila read-alignment file is a ChIP-Seq experiment from the
ENCODE project, and is available at:
https://www.encodeproject.org/files/ENCFF013TIL/@@download/ENCFF013TIL.bam

Before using the ENCODE file, please run the clean.py script, that
removes reads with chromosome names that are not present in the dm6 reference
genome we use.

The scripts jam_variants.bash and jam_peaks.bash contain all the commands 
necessary to run our compression and decompression followed by samtools variant
calling and MACS peak calling respectively. The path to the downstream programs
might have to be adjusted, but assuming the same version of the programs are used,
the running the bash script should run the entire pipeline. Different versions
of samtools might require small tweaks in the commands specifically to report
give it the name of the output files. The bam_peaks.bash and bam_variants.bash
scripts contain commands to run only the downstream analysis on the original
files.

Sources:
In both decode files, var_decode.py and peak_decode.py, we use a rotate
left function from the following website in order to aid with the creation
of our bitstring:
http://www.falatic.com/index.php/108/python-and-bitwise-rotation

Additionally, the file fasta.py, which generates the fai file and searches
for the genome substring corresponding to each read is taken from code
provided by Dr. Langmead at the following location:
https://github.com/BenLangmead/comp-genomics-class/blob/master/
notebooks/FASTA.ipynb