Filter
Exclude
Time range
-
Near
Our paper "Many Needles in a Haystack" has been accepted at ICML 2026 โ€” see you in Seoul! ๐Ÿ‡ฐ๐Ÿ‡ท๐Ÿงฌ CRISPR screens can test thousands of genes, but budgets are tight and hits are rare. Which perturbations should you run next? We frame this as a lab-in-the-loop design problem: AI proposes a batch โ†’ lab runs it โ†’ readouts update the model โ†’ repeat. Each cycle gets smarter about where hits are hiding. Our method, Probability-of-Hit, recovers more hits across 5 real immunology screens. More hits per plate, fewer wasted wells. Great work by Andrea Rubbi, Arpit Merchant, Samuel Ogden, Amir Akbarnejad, with Pietro Lio & Sattar Vakili ๐ŸŽ‰ #ICML2026 #ActiveLearning #PerturbSeq #FunctionalGenomics #CRISPRscreen #LabInTheLoop #AI4Science #BayesianOptimization
5
30
222
22,491
๊น…์ฝ” ( $DNA )์˜ ์˜๋ฏธ์žˆ๋Š” ํ–‰๋ณด๋ฅผ ํƒ€์ž„๋ผ์ธ์œผ๋กœ ์ •๋ฆฌํ–ˆ๋‹ค. โ€”โ€” 2024.02 Technology Network ์ถœ๋ฒ”. 25๊ฐœ ์ด์ƒ ํŒŒํŠธ๋„ˆ์‚ฌ์™€ AI-์‹คํ—˜ ์—ฐ๊ฒฐ ์ƒํƒœ๊ณ„ ๊ตฌ์ถ• ์„ ์–ธ. 2025.07 DOE ์‚ฐํ•˜ PNNL๊ณผ $4.66M ๊ณ„์•ฝ ์ฒด๊ฒฐ. RAC ๊ธฐ๋ฐ˜ ์ž์œจ์‹คํ—˜์‹ค(AMP2) ๊ตฌ์ถ• ์ฐฉ์ˆ˜. 2025.12 PNNL 2์ฐจ ๊ณ„์•ฝ. ์ตœ๋Œ€ $47M, 4๋…„. RAC 97๋Œ€, ๋ถ„์„ ๊ธฐ๊ธฐ 100๊ฐœ ์ด์ƒ. Genesis Mission ํ–‰์ •๋ช…๋ น ์ฒซ ๋ฒˆ์งธ ์‹คํ–‰ ์‚ฌ๋ก€. 2026.01 AMP2 ๊ณต์‹ ๊ฐ€๋™. ์ง€๊ธˆ ์ด ์ˆœ๊ฐ„์—๋„ ๋Œ์•„๊ฐ€๊ณ  ์žˆ๋‹ค. 2026.02 OpenAI ร— ๊น…์ฝ”. GPT-5 ์ž์œจ์‹คํ—˜์‹ค. 36,000๊ฐœ ์‹คํ—˜, ๋‹จ๋ฐฑ์งˆ ํ•ฉ์„ฑ ๋น„์šฉ 40% ์ ˆ๊ฐ. 2026.03 Cloud Lab ์ถœ์‹œ. ์›น ๋ธŒ๋ผ์šฐ์ €๋กœ ์ž์œจ์‹คํ—˜์‹ค ์ ‘์†. ์‹คํ—˜ 1๊ฑด $39๋ถ€ํ„ฐ. 2026.04 AWS Amazon Bio Discovery ์ถœ์‹œ. wet-lab ์ž๋™ํ™” ํŒŒํŠธ๋„ˆ๋กœ ๊น…์ฝ” ํŽธ์ž…. โ€”โ€” ๋ฏผ๊ฐ„ ๋น…ํ…Œํฌ์™€ ๋ฏธ๊ตญ ์—ฐ๋ฐฉ์ •๋ถ€๊ฐ€ ๊ฐ™์€ ์‹คํ—˜ ์ธํ”„๋ผ ์œ„์—์„œ ๋Œ์•„๊ฐ„๋‹ค. ์•„๋ฌด๋„ ์ฃผ๋ชฉํ•˜์ง€ ์•Š๊ณ  ์ธ๊ธฐ ์—†๋Š” ์ข…๋ชฉ์ด์ง€๋งŒ, ํšŒ์‚ฌ๋Š” ๋ถ€์ง€๋Ÿฐํžˆ ์•ž๊ธธ์„ ๊ฐœ์ฒ™ ์ค‘. ์ € ๋’ค๋กœ ๋” ์–ผ๋งˆ๋‚˜ ์˜๋ฏธ์žˆ๋Š” ์†Œ์‹์ด ๋“ค๋ ค์˜ฌ์ง€๐Ÿ‘€๐Ÿ‘€ #DrugDiscovery #Bioeconomy #GenesisMission #AWS #OpenAI #LabInTheLoop
2
9
40
5,492
Mar 20
Aurelien Personnaz et al.: AXIS: a Lab-in-the-Loop machine learning approach for automated detection of macromolecular crystals #StructuralBiology #AI #LabInTheLoop ... #IUCr journals.iucr.org/paper?S205โ€ฆ

2
4
379
"But again, we're seeing this feedback loop and flywheel effect. So more diagnostics means more data, and triangulating this data with the health information of the patient feeds back to the algorithm to make the diagnostics more effective. One of our portfolio holdings companies, a diagnostics company called ChinDX, believes that this combination, this data set of, in that case, genetic testing, and the data about the patients and their lineage is their main competitive advantage, their main mode." Mikhail Zverev discusses further in the short clip below ๐Ÿ‘‡ Tune into the full discussion: "Lab in the loop" and the acceleration in life sciences innovation ๐Ÿ‘‰lnkd.in/eAdFV5mi โš ๏ธ Warning: Capital at Risk. Professional Investors Only. #AmatiGlobal #innovation #healthcare #investing #investment #fundmanager #labintheloop #clinicaltrials #diagnostics #wealthmanagement #assetmanagement #fundresearch #globalequities #lifesciences
2
64
Lab-in-the-Loop drug discovery is becoming real-time and multimodal. BrainStorm Therapeutics blends organoids AI to map brain disease, repurposing an Alzheimerโ€™s drug for Rett syndrome in 9 months. The Playbook: start with human biology -> learn with AI -> validate in tight experimental loops #LabInTheLoop
2
5
39
6,225