• Authors: F. Zanini, R. Neher (+ other authors for the non-coding part), 2013-2015
• License: MIT

• Clone the repository.
• Install the dependencies:
• python2.7
• numpy
• pandas
• matplotlib
• biopython
• seqanpy (https://github.com/iosonofabio/seqanpy)
• Add the repo folder to your PYTHONPATH, so that import hivwholeseq works.
• Add the relevant data to the data folder.
• Set the environment variable HIVWHOLESEQ_ROOT_DATA_FOLDER to a folder where the read data will be stored.
• if you are using nonstandard paths, set STAMPY_BIN and FASTTREE_BIN

The analysis of HIV whole-genome longitudinal sequences consists of a set of scripts and modules. There are two levels of analysis:

1. SAMPLE: Reads from each sequenced sample must be cleaned and assigned to regions. This step is performed on non-patient samples too. The scripts and modules are located in the "sequencing" folder.

2. PATIENT: Cleaned reads are mapped against a patient-specific reference (the majotiry sequence of the first time point), allele counts and linkage data structures are precomputed. The scripts and modules are located in the subfolder "store".

Scripts exist for checking the status of the analysis, both for the SAMPLE and for the PATIENT parts; see README in the respective folders.

A standard control for sequencing quality is co-sequencing of PhiX, a plasmid of known sequence. The analysis pipeline on those reads is located in the README, "phix" folder.

See sequencing/README.

To be done after 1. See store/README.

See patients/patients.py and patients/samples.py for general classes, and the scripts in patients for typical analyses.

At this stage, the HIVEVO_access repository is recommended. The codebase is much smaller and personal data on patients/samples should be hidden.