Human Protein Complex-Based Drug Signatures for Personalized Cancer Medicine

Area of Science:

• Computational biology
• Bioinformatics
• Genomics

Background:

• Traditional drug repositioning relies on disease signatures from gene expression profiles.
• Obtaining reliable disease signatures often requires large sample sizes, which are not always feasible, especially in personalized medicine.

Purpose of the Study:

• To propose a novel drug repositioning approach (HDgS) that utilizes drug-induced gene expression profiles to identify drug signatures.
• To leverage human protein complexes (HPC) for defining drug signatures, accounting for gene dependencies.
• To enable effective drug repositioning for personalized medicine, even with minimal patient data.

Main Methods:

• The HDgS approach first identifies a drug signature from drug-induced gene expression profiles.
• Human protein complexes (HPC) are used to define the drug signature, considering inter-gene dependencies.
• The identified drug signature is then connected to disease gene expression profiles to identify potential drugs.

Main Results:

• The HPC-based drug signature effectively identifies potential drug candidates for patients.
• The HDgS method demonstrated successful application in drug repositioning for cancer samples using LINCS data.
• The approach is suitable for personalized medicine, performing well even with a single patient sample.

Conclusions:

• The proposed HDgS method offers an effective strategy for drug repositioning by utilizing HPC-based drug signatures.
• HDgS facilitates personalized medicine by enabling drug discovery with limited patient-specific gene expression data.
• This approach enhances the potential for identifying novel therapeutic agents for various diseases.