Cancer type and survival prediction based on transcriptomic feature map

Upload all source data of our paper "Cancer type and survival prediction based on transcriptomic feature map" # Zendo Project This project encompasses a series of files related to the integration of cancer survival data, feature extraction, and the augmentation of data using Generative Adversarial Networks (GANs). The project aims to improve the accuracy of survival predictions and support cancer research through in-depth analysis and processing of cancer-related data. ## Project Overview This repository includes HDF5 files of cancer survival data, intermediate files for feature extraction, and data files augmented using GAN networks. Among them, `Data Integration.ipynb` is a Jupyter notebook used for slicing and merging `feat_part1.h5` to `feat_part6.h5`, while files starting with four letters are data files for specific cancer types, and files starting with `GAN` are data that have been augmented by GAN networks. ## File Descriptions ### Data Files - BLCA.h5: Survival data for bladder cancer. - BRCA.h5: Survival data for breast cancer. - HNSC.h5: Survival data for head and neck cancer. - KIRC.h5: Survival data for kidney clear cell carcinoma. - LIHC.h5: Survival data for liver hepatocellular carcinoma. - LUAD.h5: Survival data for lung adenocarcinoma. - LUSC.h5: Survival data for lung squamous cell carcinoma. - OVxx.h5: Survival data for ovarian cancer. - SKCM.h5: Survival data for skin cutaneous melanoma. - STAD.h5: Survival data for stomach adenocarcinoma. ### Feature Extraction Files - feat_part1.h5 to feat_part6.h5: Intermediate data files for feature extraction, used to store partial feature data during the processing. ### Feature Map Files - fmap_BLCA.h5: Feature map data for bladder cancer. - fmap_BRCA.h5: Feature map data for breast cancer. - fmap_HNSC.h5: Feature map data for head and neck cancer. - fmap_KIRC.h5: Feature map data for kidney clear cell carcinoma. - fmap_LIHC.h5: Feature map data for liver hepatocellular carcinoma. - fmap_LUAD.h5: Feature map data for lung adenocarcinoma. - fmap_LUSC.h5: Feature map data for lung squamous cell carcinoma. - fmap_OVxx.h5: Feature map data for ovarian cancer. - fmap_SKCM.h5: Feature map data for skin cutaneous melanoma. - fmap_STAD.h5: Feature map data for stomach adenocarcinoma. ### GAN Augmented Data Files - GAN500sur_BLCA.h5: Augmented survival data for bladder cancer. - GAN500sur_BRCA.h5: Augmented survival data for breast cancer. - GAN500sur_HNSC.h5: Augmented survival data for head and neck cancer. - GAN500sur_KIRC.h5: Augmented survival data for kidney clear cell carcinoma. - GAN500sur_LIHC.h5: Augmented survival data for liver hepatocellular carcinoma. - GAN500sur_LUAD.h5: Augmented survival data for lung adenocarcinoma. - GAN500sur_LUSC.h5: Augmented survival data for lung squamous cell carcinoma. - GAN500sur_OVxx.h5: Augmented survival data for ovarian cancer. - GAN500sur_SKCM.h5: Augmented survival data for skin cutaneous melanoma. - GAN500sur_STAD.h5: Augmented survival data for stomach adenocarcinoma. ### Jupyter Notebook - Data Integration.ipynb: The code for slicing and merging `feat_part1.h5` to `feat_part6.h5`, a core tool in the feature integration process. ## Usage Guide 1. Data Preparation: Place the original data files in the project directory. 2. Feature Extraction: Run the `Data Integration.ipynb` notebook to merge the intermediate feature extraction files. 3. Data Augmentation: Use GAN networks to augment survival data, generating new data files. 4. Analysis Results: Use the augmented data for survival prediction analysis. ## Contributing and Collaboration We welcome contributions and collaboration in any form. If you have any questions or suggestions about the project, please contact us through GitHub Issues.