A socratic dialogue over the utility of DNA language models (Part 1 of 2)

3.7k words, 17 minute reading time

Introduction The dialogue Part 1 is focused on variant pathogenicity prediction using these models. Part 2 is focused on genome generation using these models. I am aware that DNA language models are useful for things other than those two (like protein fitness), but variant prediction and genome generation are the two bits that I find most interesting. I think I, alongside many other people in this field, live in this seemingly parallel universe where we don’t really understand why anyone is working on DNA language models. I say ‘parallel’, because there is obviously a world in which…

related reading

• A socratic dialogue over the utility of DNA language models (Part 1 of 2)owlposting.com
• A socratic dialogue over the utility of DNA language models (Part 1 of 2)owlposting.com
• A socratic dialogue over the utility of DNA language models (Part 1 of 2)substack.com
• A socratic dialogue over the utility of DNA language models (Part 2 of 2)owlposting.com
• Evo 2 Can Design Entire Genomesasimov.press
• Genome modelling and design across all domains of life with Evo 2 | Naturenature.com
• Evo 2: DNA Foundation Model | Arc Institutearcinstitute.org
• Evo 2 Can Design Entire Genomesasimov.press
• The Poetry Fan Who Taught an LLM to Read and Write DNA | Quanta Magazinequantamagazine.org
• Evo 2: DNA Foundation Model | Arc Institutearcinstitute.org
• AlphaGenome: AI for better understanding the genome — Google DeepMinddeepmind.google
• Advancing regulatory variant effect prediction with AlphaGenomenature.com