I-Corps: Novel molecule discovery by precisely predicting and modulating human pathology

这个I-Corps项目更广泛的影响/商业潜力是开发技术，加速与农业，环境，能源和人类健康/福祉相关的分子发现。该系统旨在实现对分子功能的可靠预测，其中在经典人工智能（AI）算法中没有观察到类似的数据作为训练数据。这种解决方案可以提供新的功能分子的开发，如固氮分子，可堆肥塑料，高效电池，用于分子递送的纳米笼，以及用于不治之症的安全有效的药物。此外，分布外泛化是计算机科学和数学基础研究中人工智能的一个挑战。该系统旨在展示一个可以在生命科学以外的广泛领域（如公共卫生、政治学、经济学和教育）采用的框架。该I-Corps项目开发了一个基于机器学习、生物物理学和系统生物学研究的分子发现平台。 该解决方案由下一代测序和高通量技术生成的组学数据提供动力，并由最先进的深度学习提供支持。该方法的新颖之处在于它将来自不同资源的数据整合到一个统一的框架中，以表示复杂的生物系统，并使机器学习能够生成假设。一个成果可能是新的疗法，有可能治疗复杂的疾病，如癌症，神经退行性疾病和药物滥用disorders.This奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

The broader impact/commercial potential of this I-Corps project is the development of technology accelerating molecular discovery relevant to meeting urgent challenges in agriculture, environment, energy, and human health/well-being. This system seeks to enable reliable prediction of molecule functions where no similar data has been observed as training data in classic artificial intelligence (AI) algorithms. This solution may provide for the development of new functional molecules such as nitrogen-fixing molecules, compostable plastic, high-efficiency batteries, nanocages for molecular delivery, and safe and effective medicines for incurable diseases. Furthermore, out-of-distribution generalization is a challenge in AI in fundamental research in computer science and mathematics. This system seeks to demonstrate a framework that could be adopted in a broad spectrum of fields beyond the life sciences such as public health, political science, economics, and education.This I-Corps project develops a molecular discovery platform based on studies in machine learning, biophysics, and systems biology. The solution is fueled by omics data generated by next-generation sequencing and high-throughput techniques and powered by state-of-the-art deep learning. The novelty of the approach is its way of integrating data from diverse resources into a unified framework to represent complex biological systems, and enabling machine learning for hypothesis generation. An outcome may be novel therapeutics that have the potential to treat complex diseases such as cancers, neurodegenerative diseases, and substance abuse disorders.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.