Bioinformatic workflow fragment discovery leveraging the social-aware knowledge graph.

Area of Science:

• Bioinformatics
• Computational Biology
• Knowledge Graphs

Background:

• Scientific workflow reuse saves effort and reduces errors.
• Existing methods for discovering workflow fragments focus on service invocation, neglecting social relations.
• Implicit relations between services and developers are underexplored.

Purpose of the Study:

• To propose a social-aware scientific workflow knowledge graph (S²KG).
• To capture entities and relations from bioinformatic workflows and developer information.
• To evaluate the feasibility of workflow fragment construction using S²KG.

Main Methods:

• Developed a social-aware scientific workflow knowledge graph (S²KG).
• Analyzed bioinformatic workflows and developer data from repositories.
• Extracted a service invoking network and constructed service communities.
• Implemented a fragment discovery mechanism using Yen's method.

Main Results:

• S²KG captures diverse entities and relations, including social aspects.
• The feasibility of workflow fragment construction is evaluated using S²KG attributes.
• Service communities are identified to aid single service discovery.
• The proposed method demonstrates superior precision, recall, and F1 scores compared to state-of-the-art techniques.

Conclusions:

• The social-aware approach enhances bioinformatic workflow fragment discovery.
• S²KG effectively models complex relationships within scientific workflows.
• The developed mechanism offers a more efficient and accurate solution for workflow fragment retrieval.