Network Science is a new paradigm of study to understand complex systems including biological interactions. Pharmacology and drug discovery leverage the network biology to better understand the complex interactions between drugs, targets and disease for designing new molecules or identifying repurposed drugs.
Most recently Representation Learning (RL) has found its niche for graph-structured data with state-of-the-art results for various domains. With the record of successes in network biology, there has been a tremendous surge in leveraging representation learning for biological networks from modeling to learning with networks. At its core, the spectrum of algorithmic approaches facilitated with a trained behavior over topological features convert networks into vector spaces.
The amount of applications produced by these vector space are of paramount importance as they yield to perform the tasks like predicting missing links in a graph, graph clustering, graph classification, graph generation, graph alignment, etc. Inherently, the notion of representation learning for biological networks reflect an optimized view of network topology over algorithmic paradigm of hand-curated and hard-coding feature techniques, that lack to understand the higher-order structures and often fail to embrace the inductive capability as they do not in influence network information into the predictive models. Therefore, the proposition of representation learning models for biological networks of complex design and multi-modal topological structure emerges to be a suitable direction for future advancements as this research proposes to find the efficient representations for biological networks and solve network related learning tasks. RL expressed to be a potential game changer in deciphering the inherent complex interaction patterns in a complex network more precisely.
This Research Topic aims to highlight novel research in the area of network science and network biology coupled with representation learning and its application to biology, medicine, and pharmacology. We try to focus on following broad areas but not limited to.
1. Biological Graph Representation Learning
2. Graph Convolution Neural Network for Network Biology
3. Large Biological Network Embedding
4. Heterogeneous Biological Network Integration
5. Protein Networks Alignment
5. Biological Network Module Detection
6. Prioritizing Disease Biomarkers from Complex Network
7. Inference and Analysis of Biological Network
8. Network Drug Discovery and Repurposing
Network Science is a new paradigm of study to understand complex systems including biological interactions. Pharmacology and drug discovery leverage the network biology to better understand the complex interactions between drugs, targets and disease for designing new molecules or identifying repurposed drugs.
Most recently Representation Learning (RL) has found its niche for graph-structured data with state-of-the-art results for various domains. With the record of successes in network biology, there has been a tremendous surge in leveraging representation learning for biological networks from modeling to learning with networks. At its core, the spectrum of algorithmic approaches facilitated with a trained behavior over topological features convert networks into vector spaces.
The amount of applications produced by these vector space are of paramount importance as they yield to perform the tasks like predicting missing links in a graph, graph clustering, graph classification, graph generation, graph alignment, etc. Inherently, the notion of representation learning for biological networks reflect an optimized view of network topology over algorithmic paradigm of hand-curated and hard-coding feature techniques, that lack to understand the higher-order structures and often fail to embrace the inductive capability as they do not in influence network information into the predictive models. Therefore, the proposition of representation learning models for biological networks of complex design and multi-modal topological structure emerges to be a suitable direction for future advancements as this research proposes to find the efficient representations for biological networks and solve network related learning tasks. RL expressed to be a potential game changer in deciphering the inherent complex interaction patterns in a complex network more precisely.
This Research Topic aims to highlight novel research in the area of network science and network biology coupled with representation learning and its application to biology, medicine, and pharmacology. We try to focus on following broad areas but not limited to.
1. Biological Graph Representation Learning
2. Graph Convolution Neural Network for Network Biology
3. Large Biological Network Embedding
4. Heterogeneous Biological Network Integration
5. Protein Networks Alignment
5. Biological Network Module Detection
6. Prioritizing Disease Biomarkers from Complex Network
7. Inference and Analysis of Biological Network
8. Network Drug Discovery and Repurposing