空气中二氧化碳(Carbon Dioxide, CO2)浓度逐步上升，全球变暖问题日益严重。目前全球已有120多个国家提出了碳中和目标，碳捕集、碳封存和碳转化技术受到学术界和工业界的高度重视。生物固碳技术因反应条件温和、绿色环保及反应产品应用范围广等特点，在碳中和背景下显示出良好的工业应用潜力。然而，要真正实现可持续发展，必须采用有效评价策略用以量化生物固碳效益。但目前的生物固碳效益评价方法多种多样，缺乏统一的科学评估框架。为了促进生物固碳效益评价的发展，本文综述了基因组规模代谢网络模型(Genome-Scale Metabolic Network Model, GSM)和生命周期评价(Life Cycle Assessment, LCA)在生物固碳评价中的应用，重点分析了这两种方法在生物固碳效益评价中的瓶颈问题。此外，本综述还对生物固碳效益评价的未来发展方向进行了系统展望，阐述了机器学习应用的显著优势，指出了基于原子经济性(Atomic Economy, AE)的科学评估指标，规范了数据获取的基本框架，并分析了构建多层次评估框架的重要突破。

With the concentration of carbon dioxide (CO2) in the air rising, the threat of global warming is getting worse. Due to its mild conditions and reaction specificity, the biological carbon sequestration technology shows excellent potential for industrial applications under carbon neutral constraints. However, to be genuinely sustainable, industrial applications must implement low-carbon-footprint technology. Current methods for biological carbon sequestration assessment are few and heterogeneous, with the lack of a harmonized scientific assessment framework. In order to promote the development on the biological carbon sequestration assessment, this article reviews the application of genome-scale metabolic network model (GSM) and life cycle assessment (LCA) in biological carbon sequestration assessment, with a focus on analyzing the bottleneck issues of these two methods in biological carbon sequestration assessment. In addition, this review also provides a systematic outlook on the future development direction of the biological carbon sequestration assessment, elaborates on the significant advantages of machine learning applications, points out the scientificity of evaluation indicators based on atomic economy (AE), standardizes the basic framework for data acquisition, and analyzes important breakthroughs in constructing a multi-level evaluation framework.