Scholarlink

<p>The development of organic sensitizers with high molar extinction (<i>ε</i>) coefficients is important for various light absorption applications. To accelerate the discovery of such compounds, a machine learning (<i>ML</i>) analysis has been applied to explore their vast chemical space. A dataset of 676 organic chromophores is analyzed by designing their electronic, topological, and molecular descriptors to predict their <i>ε</i>. Among the 10 tested <i>ML</i> models, Gradient Boosting, Random Forest, Extra Trees, and Historical Gradient Boosting regressors show good correlation with their experimental and predicted values (<i>R</i><sup><i>2</i></sup> ≈ 0.70). Their Shapley Feature importance reveals that Subgraph Density of Secondary Carbon-Hydrogen (<i>SdsCH</i>) and logarithm of the partition coefficient- an Der Waals Surface Area Descriptor 8 (<i>SlogP_VSA8</i>) Descriptors have a significant impact on model performance. Additionally, by leveraging breaking retrosynthetic analysis, 3288 novel structures with potential high <i>ε</i> have been synthesized to validate their feasibility through dimensionality reduction analysis. Their synthetic accessibility (<i>SA</i>) calculations identify the top structures for their experimental synthesis in the future. Interestingly, the findings indicate that new structures with <i>SMILES</i> lengths of 35–80 units can exhibit the highest <i>SA</i>.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Accelerated Chemical Space Generation of High Molar Extinction Organic Sensitizers via Machine Learning

  • 期刊论文

  • 发表日期：

    2025年9月18日

查看全文
  • Sadaf Noreen,
  • Mamduh J. Aljaafreh

摘要

The development of organic sensitizers with high molar extinction (ε) coefficients is important for various light absorption applications. To accelerate the discovery of such compounds, a machine learning (ML) analysis has been applied to explore their vast chemical space. A dataset of 676 organic chromophores is analyzed by designing their electronic, topological, and molecular descriptors to predict their ε. Among the 10 tested ML models, Gradient Boosting, Random Forest, Extra Trees, and Historical Gradient Boosting regressors show good correlation with their experimental and predicted values (R2 ≈ 0.70). Their Shapley Feature importance reveals that Subgraph Density of Secondary Carbon-Hydrogen (SdsCH) and logarithm of the partition coefficient- an Der Waals Surface Area Descriptor 8 (SlogP_VSA8) Descriptors have a significant impact on model performance. Additionally, by leveraging breaking retrosynthetic analysis, 3288 novel structures with potential high ε have been synthesized to validate their feasibility through dimensionality reduction analysis. Their synthetic accessibility (SA) calculations identify the top structures for their experimental synthesis in the future. Interestingly, the findings indicate that new structures with SMILES lengths of 35–80 units can exhibit the highest SA.