RegenX: an NLP recommendation engine for neuroregeneration topics over time

Area of Science:

• Neuroscience
• Bioinformatics
• Computational Biology

Background:

• The regeneration of central nervous system (CNS) neurons and axons is a significant challenge in neuroscience.
• Historical literature (pre-1990) offers insights into conditions promoting nerve regeneration, lacking molecular detail.
• Post-1990 advancements, including omics technologies, have generated vast datasets crucial for neuroregeneration research.

Purpose of the Study:

• To explore the evolution of neuroregeneration literature from the 1700s to the present.
• To apply natural language processing (NLP) techniques for extracting actionable intelligence from historical and modern research.
• To develop tools for researchers to efficiently access and understand domain-specific literature.

Main Methods:

• Curated over 600 published works in neuroregeneration.
• Applied dynamic topic modeling using the Latent Dirichlet allocation (LDA) algorithm to cluster research topics.
• Developed a recommendation engine and interactive visualization interface.

Main Results:

• Identified how research topics in neuroregeneration cluster and evolve over time.
• Created a recommendation engine that matches researcher queries to relevant document topics.
• Provided interactive visualizations for exploring topic dynamics and composition changes.

Conclusions:

• Introduced a novel recommendation engine and interactive interface for dynamic topic modeling in neuronal regeneration.
• Facilitated easier access to domain-specific literature for researchers.
• Enabled deeper understanding of the historical and current landscape of neuroregeneration research.