Quantum enhanced stratification of Breast Cancer: exploring quantum expressivity for real omics data

Stratification of tumour samples from molecular descriptors (gene expression and copy number alteration) with quantum Kernels

Input data

Input data contains the METABRIC dataset reduced with UMAP and the notebook to see the dimentionality recuction step.

Experimental Results

Kernel_Results

This folder contains all quantum kernel computed in this work

Results

This folder contains clustering results extracted for all the computed kernels

Packages Requirments and usage

To avoide problems with the env requirments follow these steps:

1. Create conda env with python 3.10

conda create -n <your_env_name> python==3.10

2. Activate env

conda activate <your_env_name>

3. Install all packages present in requirments.txt

pip install -r requirments.txt

Usage

Launch noisless Quantum Kernel simulation:

python Qkernel_comp_unsup_simulation.py -params utils/hyper_param_unsup.json 

Launch QPU Quantum Kernel computation:

python Qkernel_real_hardware_CU_tr.py -params utils/Qkernel_real_hardware_CU_tr.py

Launch clustering of a given set of kernels and compute Silhouette scores

python Analysis_unsup.py -params utils/hyper_param_unsup_analysis.json