Alpha Lee's guest appearances

Alpha Lee

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

Out of the Crisis

Bold Conjectures with Paras Chopra

Michael Levin

Jeremy Howard

Meredith Broussard

Ilias Diakonikolas

Scott Clark

Celeste Kidd

Brian Armstrong

Dennis Mortensen

Rob Walker

Sal Khan

Stephan Guyenet

Sanmi Koyejo

Chelsea Finn

Dario Amodei

Jianxiong Xiao

OwlTail

Can you crowdsource your way to discovering a new drug? Alpha Lee's startup PostEra launched a crowdsourced effort (called Covid Moonshot) to discover drugs that are effective in treating covid-19. This was the first of its kind effort that combined research efforts from several groups across the world to design a drug that treats covid-19, is free from patents and can be made cheaply and easily anywhere in the world. == What we talk about == 0:04 - Introduction 1:28 - What does your company, PostEra, do? 4:58 - Why is the traditional method of discovering drugs inefficient and hard? 9:52 - How do you find whether a computationally discovered drug will work? 11:25 - How do you optimize the drugs for toxicity, solubility, and other properties? 12:03 - How are you able to come up with a chemical synthesis pathway in a finite time? 16:44 - What are the areas where deep learning models have become better than humans in drug development? 19:18 - What’s the origin story of Covid Moonshot? How did it grow into what it is now? 25:24 - How was the experience managing people from around the world working on this project, especially when there is no central authority? 27:47 - The process of finding the target molecule 34:17 - Since Covid moonshot is all pro-bono and you are not filing for a patent, how do you see the process of regulatory trials, which is very costly, being done? 37:19 - By the time your drug comes, do you think the Covid-19 will be a much lesser threat? 39-49 - Do you plan to use the same approach for different diseases? And do you also envision driving the whole process of drug discovery in the future, and not just enablement? 41:41 - How do you align your science research interests with commercial interests? 43:46 - Haș an insight, while working on a problem, inspired a scientific question? 45:23 - The marriage of use of basic science and real-world problems 46:38 - How can someone help in the Covid Moonshot mission? == About the guest == Dr Alpha Lee is the co-founder of PostEra, a start-up that offers medicinal chemistry as a service powered by machine learning. He did his Ph.D. at the University of Oxford. Currently, he’s at the University of Cambridge where his research group explores the integration of physics, statistics, and machine learning. 

Can you crowdsource your way to discovering a new drug? Alpha Lee's startup PostEra launched a crowdsourced effort (called Covid Moonshot) to discover drugs that are effective in treating covid-19. This was the first of its kind effort that combined research efforts from several groups across the world to design a drug that treats covid-19, is free from patents and can be made cheaply and easily anywhere in the world. == What we talk about == <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=4s" rel="nofollow" target="_blank">0:04</a> - Introduction <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=88s" rel="nofollow" target="_blank">1:28</a> - What does your company, PostEra, do? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=298s" rel="nofollow" target="_blank">4:58</a> - Why is the traditional method of discovering drugs inefficient and hard? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=592s" rel="nofollow" target="_blank">9:52</a> - How do you find whether a computationally discovered drug will work? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=685s" rel="nofollow" target="_blank">11:25</a> - How do you optimize the drugs for toxicity, solubility, and other properties? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=723s" rel="nofollow" target="_blank">12:03</a> - How are you able to come up with a chemical synthesis pathway in a finite time? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=1004s" rel="nofollow" target="_blank">16:44</a> - What are the areas where deep learning models have become better than humans in drug development? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=1158s" rel="nofollow" target="_blank">19:18</a> - What’s the origin story of Covid Moonshot? How did it grow into what it is now? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=1524s" rel="nofollow" target="_blank">25:24</a> - How was the experience managing people from around the world working on this project, especially when there is no central authority? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=1667s" rel="nofollow" target="_blank">27:47</a> - The process of finding the target molecule <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2057s" rel="nofollow" target="_blank">34:17</a> - Since Covid moonshot is all pro-bono and you are not filing for a patent, how do you see the process of regulatory trials, which is very costly, being done? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2239s" rel="nofollow" target="_blank">37:19</a> - By the time your drug comes, do you think the Covid-19 will be a much lesser threat? 39-49 - Do you plan to use the same approach for different diseases? And do you also envision driving the whole process of drug discovery in the future, and not just enablement? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2501s" rel="nofollow" target="_blank">41:41</a> - How do you align your science research interests with commercial interests? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2626s" rel="nofollow" target="_blank">43:46</a> - Haș an insight, while working on a problem, inspired a scientific question? <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2723s" rel="nofollow" target="_blank">45:23</a> - The marriage of use of basic science and real-world problems <a href="https://www.youtube.com/watch?v=-F55ffPDjgc&amp;t=2798s" rel="nofollow" target="_blank">46:38</a> - How can someone help in the Covid Moonshot mission? == About the guest == Dr Alpha Lee is the co-founder of PostEra, a start-up that offers medicinal chemistry as a service powered by machine learning. He did his Ph.D. at the University of Oxford. Currently, he’s at the University of Cambridge where his research group explores the integration of physics, statistics, and machine learning.

#17 Alpha Lee - Crowdsourcing Discovery of a COVID-19 Drug

After seeing the impact of the SARS pandemic as a child in Hong Kong, Alpha Lee dedicated his life to science. Studying mathematics and biology at some of the most prestigious universities in the world, Alpha has now turned his attention to an area ripe for disruption, preclinical drug discovery.Alpha had just arrived in the bay area for the startup program Y Combinator when the lockdowns for COVID starting. Having been through this before, Alpha knew what he needed to do. Alpha decided to use the technology he had been developing to find promising compounds to fight coronavirus and then open source it. Alpha received thousands of submissions and now the most promising of those are being developed in a lab in the UK. Alpha's message throughout all of this? Let's not waste the opportunity, this time.Resources we discussed:https://postera.ai/covidhttps://www.alpha-lee.com/https://www.gofundme.com/f/covidmoonshothttps://covid19-hpc-consortium.org/projects/5e8be82df1a9290078584cf4

After seeing the impact of the SARS pandemic as a child in Hong Kong, Alpha Lee dedicated his life to science. Studying mathematics and biology at some of the most prestigious universities in the world, Alpha has now turned his attention to an area ripe for disruption, preclinical drug discovery.Alpha had just arrived in the bay area for the startup program Y Combinator when the lockdowns for COVID starting. Having been through this before, Alpha knew what he needed to do. Alpha decided to use the technology he had been developing to find promising compounds to fight coronavirus and then open source it. Alpha received thousands of submissions and now the most promising of those are being developed in a lab in the UK. Alpha's message throughout all of this? Let's not waste the opportunity, this time.Resources we discussed:<a href="https://postera.ai/covid" rel="nofollow" target="_blank">https://postera.ai/covid</a><a href="https://www.alpha-lee.com/" rel="nofollow" target="_blank">https://www.alpha-lee.com/</a><a href="https://www.gofundme.com/f/covidmoonshot" rel="nofollow" target="_blank">https://www.gofundme.com/f/covidmoonshot</a><a href="https://covid19-hpc-consortium.org/projects/5e8be82df1a9290078584cf4" rel="nofollow" target="_blank">https://covid19-hpc-consortium.org/projects/5e8be82df1a9290078584cf4</a>

Alpha Lee: An Open Source COVID Moonshot

Today we’re joined by Alpha Lee, Winton Advanced Fellow in the Department of Physics at the University of Cambridge. Our conversation centers around Alpha’s research which can be broken down into three main categories: data-driven drug discovery, material discovery, and physical analysis of machine learning. We discuss the similarities and differences between drug discovery and material science, his startup, PostEra which offers medicinal chemistry as a service powered by machine learning, and much more

The Physics of Data with Alpha Lee - #377