CDI-Type I: Modeling Quantum Tunnel Current to Statistically Sequence Biomolecules

这个范式转换项目提高了阅读DNA和蛋白质的能力，并建立在使用机器学习设计纳米技术系统的基础上。其目标是使用计算思维来改变生物分子的设计方式，通过提供物理模型、计算数据和适应的统计学习方法的协同组合来主动考虑随机性。目前，DNA和蛋白质的全电子测序是一个投机性的梦想，可能会给生化科学和医学带来革命性的变化。这个项目从理论和算法上研究了使用计算框架的全电子测序的基本原理。目前，机器学习领域认为从数据中提取有用的特征在很大程度上独立于作用于所提取的特征的学习算法的设计。这个项目的目的是发现是否可以通过设计学习算法来获得显著的性能收益，该算法明确地考虑到电子测序中的物理变化和随机性。潜在的计算思维可能会带来洞察力和方法，从而在生物分子全电子测序方面取得革命性的进展。该项目的研究方向可以影响生物、医学、纳米电子学和机器学习系统的设计。对高中生、本科生和研究生在设计和理解纳米尺度和机器学习系统方面的量子设备和统计理论和方法的培训是该项目所涵盖活动的组成部分。该奖项是网络驱动的发现和创新计划的一部分，获奖者是华盛顿大学的M.P.Anantram教授和Maya Gupta教授。

This paradigm shifting project advances the ability to read DNA and proteins and builds on designing nanotechnology systems using machine learning. The goal is to use computational thinking for changing the way biomolecules are designed by providing a synergistic combination of physical modeling, computational data, and adapted statistical learning methods that proactively account for the randomness. Currently all-electronic sequencing of DNA and proteins is a speculative dream that could revolutionize biochemical sciences and medicine. This project investigates theoretically and algorithmically the fundamentals of all-electronic sequencing using a computational framework. Currently, the field of machine learning treats the extraction of useful features from data as largely independent from the design of learning algorithms that act on the extracted features. This project aims at discovering whether significant performance gains can be obtained by designing learning algorithms that explicitly account for the physical variations and randomness in electronic sequencing. The underlying computational thinking can potentially lead to insights and methods leading to transformative advances on all-electronic sequencing of biomolecules. The research direction of this project can impact biology, medicine, nanoelectronics, and the design of machine learning systems. The training of high school, undergraduate and graduate students in quantum devices and statistical theory and methods in the design and understanding of nanoscale and machine learning systems are an integral part of the activities spanned by this project.This award is part of the Cyber-Enabled Discovery and Innovation program, and the recipients are Professors M. P. Anantram and Maya Gupta of the University of Washington.