SRHgroup / IMPROVE_tool

IMPROVE tool

A feature based random forest model to priotize neoepitopes. This repository is bulid for tool usage. Look at all scripts made for the paper at: https://github.com/SRHgroup/IMPROVE_paper (will be made public upon acceptance)

Improve has two steps:

1. Feature_calculations.py
2. Preidct_immunogenicity.py

1. Feature calculations

Feature_calculations.py calculates all features used for the simple model besides:

NetMHCExp and Expression: These two features can be obtained by this tool:

Forginess score: Can be obtained by https://github.com/andrewrech/antigen.garnish

IMPROVE dependency and installation

Save all programs to a folder

IMPROVE Simple model

PRIME

https://github.com/GfellerLab/PRIME

Prime simple depends on:

https://github.com/GfellerLab/MixMHCpred

NetMHCpan

https://services.healthtech.dtu.dk/services/NetMHCpan-4.1/

NetMHCstaboan

https://services.healthtech.dtu.dk/services/NetMHCstabpan-1.0/

Self similarity with kernel self similarity

https://github.com/SRHgroup/self_similarity

Save all tools to a folder used for the feature calculation "--ProgramDir"

IMPROVE TME excluded

PrioScore: Priority score from Mupexi: https://services.healthtech.dtu.dk/services/MuPeXI-1.1/

CelPrev: Obtained from PyClone: https://github.com/Roth-Lab/pyclone

IMPROVE TME included (RNA sequencing data is needed)

CYT: Geometric mean of GZMA and PRF1

MCPmean: Mean score of the 10 cell populations estimated from MCP-Counter

2. Predict immunogenicity

You can predit with three diffrent versions:

1. Simple: All features from simple model and addition of NetMHCExp, Expression and Foreginess score
2. TME excluded: All features from simple model with addition of priority score and cellular prevelance
3. TME included: All features from TME excluded including RNA feature: CYT (geometirc mean of GZMA and PRF1) and MCP-mean gained from MCP-counter.

Usage

Feature calculation tool

python3 bin/feature_calculations.py --file data/test_file_for_feature_calculation_small.tsv --dataset "testing" --ProgramDir "/Users/annieborch/Documents/programs/" --outfile "data/calculated_features_test.tsv" --TmpDir "/Users/annieborch/Documents/programs/"

Predict tools:

python3 bin/Predict_immunogenicity.py --file data/calculated_features_test.tsv --model TME_excluded --outfile "output_prediction_tme_excluded.tsv"

python3 bin/Predict_immunogenicity.py --file data/calculated_features_test.tsv --model TME_included --outfile "output_prediction_tme_included.tsv"

python3 bin/Predict_immunogenicity.py --file data/calculated_features_test.tsv --model Simple --outfile "output_prediction_simple.tsv"