I would like to thank the authors for their engaging work entitled “Machine Learning of Metal-Organic Framework Design for Carbon Dioxide Capture and Utilization.” However, I have a few questions and comments regarding the study. Please note that the following remarks are intended solely to foster constructive discussion and, hopefully, contribute to advancements in the field:
The manuscript provides a comprehensive overview of machine learning (ML) applications in the design of metal-organic frameworks (MOFs). However, several critical aspects raise concerns about the robustness and real-world applicability of the findings. While the article references several databases, including CoRE MOF 2019 and ARC-MOF, there is no clear workflow or supplementary data provided to enable reproducibility of the discussed analyses. This lack of transparency undermines the reliability of the conclusions.
Furthermore, the discussion of generative models for MOF design, while intriguing, remains unsubstantiated without empirical synthesis or real-world validation of the suggested MOFs. Claims regarding the efficiency and synthesizability of these models are presented without sufficient experimental support, limiting their practical relevance. A comparative analysis of these generative models against traditional approaches would further substantiate their utility.
The manuscript also raises concerns regarding citation accuracy and misrepresentation of referenced works. For instance, the discussion of results by Bailey et al. and Gheytanzadeh et al. appears incomplete, with key context and limitations of the studies omitted. This could mislead readers regarding the robustness and significance of the cited findings.
Finally, the study lists significant challenges, such as biases in datasets and the interpretability of ML models, but fails to provide actionable strategies to address these limitations. Additionally, some sections, such as the review of ML workflows and database summaries, are verbose and repetitive, which detracts from the overall clarity and focus of the manuscript.
Looking forward to your thoughts on these points.