한 개인이 AI를 활용해 시한부 반려견의 암 치료제를 직접 만든 사례 > 호주의 한 기술직 남성 > 시한부 암에 걸린 유기견을 입양함 > 3,000달러(약 400만 원)를 들여 반려견의 종양 DNA 서열을 분석함 > 본인은 생물학 지식이 전혀 없음 > ChatGPT와 AlphaFold(단백질 구조 예측 AI)에 데이터를 입력함 > 변이 단백질을 식별하고 약물 타겟을 찾아냄 > 맨땅에서 맞춤형 mRNA 암 백신을 직접 설계함 > 유전학 교수가 "일반인이 혼자 이걸 해냈다고?"라며 경악함 > 백신 설계보다 행정 승인 절차가 더 오래 걸림 > 3개월 만에 승인받아 첫 접종 실시 > 종양 크기가 절반으로 줄고 털에 윤기가 돌아오며 건강을 회복함 > 교수: "개에게 가능하다면, 왜 인간에게는 이걸 적용하지 않고 있는가?" 단 한 명의 인간과 챗봇, 그리고 3,000달러가 거대 제약 산업의 신약 개발 시스템 전체를 압도함 AI를 통해 수많은 질병을 정복할 수 있는 시대가 이미 우리 곁에 와 있음을 보여준다.

Cloud labs make a ton of sense given the huge upfront cost to start a bio lab (often $10M in rent, equipment, regulatory, etc) -- it would be faster and cheaper if biotech startups could generate the data they need by just ordering their experiments online. We're following cloud lab pioneers like Emerald Cloud Labs and Transcriptic/Strateos, and have a few rules for Ginkgo Cloud Lab to make it easier to reliably deliver for customers: 1. Data in / data out -- You tell us the experiment and we do the whole thing from building DNA to running the assay. We aren't taking customer samples or sending out samples yet. 2. We are initially offering "Ginkgo certified protocols" - this means ones that have gone through dry run, wet run, and biovalidation on our autonomous lab in Boston. 3. We will aggressively add to the list of certified protocols (new additions weekly) - so tell us what you want!! 4. We have an agent that will do it's best to evaluate if we can run your protocol and what it would cost. Try it here! cloud.ginkgo.bio/protocols 5. Cloud Labs will eventually allow everyone to be a scientist like personal computers let everyone be a programmer. So we will have some fun protocols too - like doing glowing bacterial pixel art :) 6. The end state will be you can run a one-off protocol that you specify on the web as long as you are willing to pay for reagent and equipment time -- help us get there by telling us what protocols you'd like to see! Please give it a try and let us know what protocols you'd like to see added next!