JessEV is a framework for simultaneous selection of epitopes and design of spacers for string-of-beads vaccines based on mixed integer linear programming. The linear program maximizes the immunogenicity of the epitopes selected, while respecting constraints related to their pathogen/HLA coverage and conservation, as well as cleavage scores at certain critical locations of the vaccine.
The required dependencies can be installed via conda:
conda create -n jessev --file packages.txt -c conda-forge
Additionally, you should install one of the solvers that are supported by pyomo. The default solver used by this project is gurobi, which is free for academic usage.
The command line interface is contained in design.py, whose usage can be seen via python design.py --help. As input it requires a CSV file with the epitopes to consider, the output file for the vaccine, and a series of options to specify constraints on the vaccine:
Usage: design.py [OPTIONS] INPUT_EPITOPES OUTPUT_VACCINE
Options:
-s, --min-spacer-length INTEGER
Minimum length of the spacer to be designed
-S, --max-spacer-length INTEGER
Maximum length of the spacer to be designed
-e, --num-epitopes INTEGER Number of epitopes in the vaccine
--top-immunogen FLOAT Only consider the top epitopes by
immunogenicity
--top-proteins FLOAT Only consider the top epitopes by protein
coverage
--top-alleles FLOAT Only consider the top epitopes by allele
coverage
--min-alleles FLOAT Vaccine must cover at least this many
alleles
--min-proteins FLOAT Vaccine must cover at least this many
proteins
--min-avg-prot-conservation FLOAT
On average, epitopes in the vaccine must
cover at least this many proteins
--min-avg-alle-conservation FLOAT
On average, epitopes in the vaccine must
cover at least this many alleles
-g, --min-nterminus-gap FLOAT Minimum cleavage gap
-n, --min-nterminus-cleavage FLOAT
Minimum cleavage at the n-terminus
-ct, --min-cterminus-cleavage FLOAT
Minimum cleavage at the n-terminus
-c, --min-spacer-cleavage FLOAT
Minimum cleavage inside the spacers
-C, --max-spacer-cleavage FLOAT
Maximum cleavage inside the spacers
-E, --max-epitope-cleavage FLOAT
Maximum cleavage inside epitopes
-i, --epitope-cleavage-ignore-first INTEGER
Ignore first amino acids for epitope
cleavage
--log-file PATH Where to save the logs
--verbose Print debug messages
--solver-type TEXT Which linear programming solver to use
--help Show this message and exit.
The input epitopes must be in a CSV file with the following columns:
immunogen: the immunogenicity of the epitope.alleles: a list of alleles to which the epitope binds separated by;, e.g.HLA-B*40:06;HLA-A*01:01;HLA-B*40:01.proteins: a list of numerical IDs of the proteins that contain the epitope, e.g.53;63;2.epitope: the epitope sequence, e.g.MGNKWSKSI.
The experiments performed in the paper can be run with the bash scripts in the experiments directory. The necessary input data is located in the dev directory, where results and log files will be placed. These results can be analyzed by running plots.py, which will create the paper's figures in the dev directory and print the results of the analyses mentioned in the paper.
Note: the sequential approach requires FRED-2, which does not officially support python 3 yet. Unfortunately, some fiddling is required on your part.