GraphCDD

This code is the implementation of GraphCDD

GraphCDD: Predicting circRNA-drug resistance associations based on a multimodal graph representation learning framework


Liu Ziqiang, Dai Qiguo(导师), et al. Predicting circRNA-drug resistance associations based on a multimodal graph representation learning framework[J]. IEEE Journal of Biomedical and Health Informatics.
DOI: 10.1109/jbhi.2023.3299423
中科院 SCI分区（2022升级版） 工程技术 1 区，JCR Q1，影响因子：7.7，TOP 期刊

Requirements

• python (tested on version 3.7.11)
• pytorch (tested on version 1.6.0)
• torch-geometric (tested on version 2.0.2)
• numpy (tested on version 1.21.2)
• scikit-learn(tested on version 1.0.2)

Quick start

To reproduce our results:
Run code\main.py to RUN GraphCDD.

Folder

• code: Model code of GraphCDD.
• datasets: Data required by GraphCDD.
• results: Results of GraphCDD run.

Data description

• allpair.csv: all pairs of circRNAs and drug resistance
• circ4.csv: circRNA integrated similarity
• CSS.csv: circRNA sequence similarity
• dis.csv: disease integrated similarity
• DSS.csv: disease semantic similarity
• drug.csv: drug integrated similarity
• MTS.csv: molecular structure similarity
• circ_dis.csv: circRNA-disease association matrix
• circ_drug.csv: circRNA-drug resistance association matrix
• drug_dis.csv: disease-drug resistance association matrix
• circRNAname.xlsx: list of circRNA names
• drugname.xlsx: list of drug names
• diseasename.xlsx: list of disease names
• NoncoRNA.xls: data used in the casestudy

Contacts

If you have any questions or comments, please feel free to email Ziqiang Liu(liuzq_dlmu@163.com)