The discovery of novel drug targets is critical and a challenging step in drug development, especially for complex diseases caused by a combined effect of genetic, environmental and lifestyle factors such as cancer. Target discovery, which involves the identification and validation of disease related macromolecule (such as proteins and nucleic acid), is currently biochemistry- and pharmacology-driven by screening and validating numerous number of compounds. However, due to the tremendous sequencing data, complicated protein 3D structures, hidden physicochemical characteristics, and so on, the traditional methods have become time-consuming, laborious and inefficient, leading to the slow process of drug development. Therefore, the comprehensive understanding of the target druggability is required.
With the development of Artificial Intelligence (AI), it has been acknowledged that AI can facilitate the extraction of key information from big data of a given complex problem. In pharmaceutical sciences, target discovery based on pre- and clinical data including massive image, sequence or molecular information has now emerged to be one of the most popular and active directions for the application of AI technique. As known, the existing AI techniques do not fit well with current pharmaceutical researches because of their poor ability in deriving more predictive hypotheses from existing data and therefore, the application of advanced AI techniques and development of pharmacy-specific AI algorithms that are important in modern drug development are urgently needed.
Moreover, the discovery of novel targets is an important and ongoing need for precision or personalized medicine of cancer and can be greatly facilitated by the integration of AI. In other words, it is necessary to further optimize and redevelop the existing AI algorithms to adapt to the latest research of target discovery for anticancer therapy, from biomarker identification, to ‘hits’ finding, lead compounds optimization to pre-clinical evaluation of drug-candidates.
Hence, we are looking forward to original studies or latest Reviews discussing the following two aspects: (1) AI-aided discovery of novel drug targets for anticancer therapy, and (2) development of AI-based algorithm in oncology pharmaceutics. Our goal is to present the developments in the field of target discovery using AI-based techniques such as machine learning (ML) and molecular dynamics (MD) simulation, that can advance our knowledge in cancer.
Specific research areas applied to target discovery for anticancer therapy included in this Research Topic (both Review and Original Research):
- Omics/multi-omics analyses and biomarker identification
- Characterization and determination of target druggability
- Accumulation of large-scale information and database development
- Construction of online tools for analysis, visualization or prediction
- New feature representation strategies for small molecules, proteins, etc.
- Integration of pharmaceutical data for achieving BIG data analysis
- Problem-based optimization of deep learning architecture
- Enhancing algorithms’ reproducibility among multiple experiments
- Algorithm development for the pharmaceutical data of small size
- High-throughput virtual screening of drug candidates
- Drug-target interaction by molecular dynamics and machine learning
The discovery of novel drug targets is critical and a challenging step in drug development, especially for complex diseases caused by a combined effect of genetic, environmental and lifestyle factors such as cancer. Target discovery, which involves the identification and validation of disease related macromolecule (such as proteins and nucleic acid), is currently biochemistry- and pharmacology-driven by screening and validating numerous number of compounds. However, due to the tremendous sequencing data, complicated protein 3D structures, hidden physicochemical characteristics, and so on, the traditional methods have become time-consuming, laborious and inefficient, leading to the slow process of drug development. Therefore, the comprehensive understanding of the target druggability is required.
With the development of Artificial Intelligence (AI), it has been acknowledged that AI can facilitate the extraction of key information from big data of a given complex problem. In pharmaceutical sciences, target discovery based on pre- and clinical data including massive image, sequence or molecular information has now emerged to be one of the most popular and active directions for the application of AI technique. As known, the existing AI techniques do not fit well with current pharmaceutical researches because of their poor ability in deriving more predictive hypotheses from existing data and therefore, the application of advanced AI techniques and development of pharmacy-specific AI algorithms that are important in modern drug development are urgently needed.
Moreover, the discovery of novel targets is an important and ongoing need for precision or personalized medicine of cancer and can be greatly facilitated by the integration of AI. In other words, it is necessary to further optimize and redevelop the existing AI algorithms to adapt to the latest research of target discovery for anticancer therapy, from biomarker identification, to ‘hits’ finding, lead compounds optimization to pre-clinical evaluation of drug-candidates.
Hence, we are looking forward to original studies or latest Reviews discussing the following two aspects: (1) AI-aided discovery of novel drug targets for anticancer therapy, and (2) development of AI-based algorithm in oncology pharmaceutics. Our goal is to present the developments in the field of target discovery using AI-based techniques such as machine learning (ML) and molecular dynamics (MD) simulation, that can advance our knowledge in cancer.
Specific research areas applied to target discovery for anticancer therapy included in this Research Topic (both Review and Original Research):
- Omics/multi-omics analyses and biomarker identification
- Characterization and determination of target druggability
- Accumulation of large-scale information and database development
- Construction of online tools for analysis, visualization or prediction
- New feature representation strategies for small molecules, proteins, etc.
- Integration of pharmaceutical data for achieving BIG data analysis
- Problem-based optimization of deep learning architecture
- Enhancing algorithms’ reproducibility among multiple experiments
- Algorithm development for the pharmaceutical data of small size
- High-throughput virtual screening of drug candidates
- Drug-target interaction by molecular dynamics and machine learning