Assessing Public Opinion on CRISPR-Cas9: Combining Crowdsourcing and Deep Learning

Area of Science:

• Bioengineering
• Genetics
• Bioethics

Background:

• CRISPR-Cas9 gene editing technology presents significant potential and risks, fueling public debate.
• Rapid technological advancement necessitates timely analysis of public opinion.
• Understanding public perception is crucial for future policy decisions.

Purpose of the Study:

• To analyze public discourse on CRISPR gene editing using social media data.
• To track the evolution of public sentiment and trending topics related to CRISPR.
• To leverage machine learning for real-time analysis of public opinion.

Main Methods:

• Utilized Bidirectional Encoder Representations from Transformers (BERT) and statistical methods.
• Analyzed all available tweets mentioning CRISPR since 2013.
• Employed natural language processing for sentiment and topic analysis.

Main Results:

• Initial tweet sentiment towards CRISPR was highly positive, decreasing over time.
• Sentiment decline was linked to specific negative events and strong negative opinions.
• Analysis revealed shifts in trending topics and their association with real-world events.

Conclusions:

• Social media analysis, particularly Twitter data, provides valuable insights into public debate on gene editing.
• BERT-based analysis offers a novel approach to empirical bioethics.
• Real-time monitoring of public discourse is essential for understanding the impact of scientific advancements.