Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.co/IYexnQd1oj
Highly accurate protein structure prediction with AlphaFold.
Mentions
Extended reach
Total likes
Video views
Mentions per month
Jul 2021 – Nov 2025Where it's discussed
-
News281 articles
-
X / Twitter222 posts
-
Blogs71 articles
-
Facebook10 posts
-
Reddit10 posts
-
YouTube5 videos
-
Bluesky4 posts
AI synopsis
What people are saying
The discussions consistently praise AlphaFold as a monumental scientific breakthrough, with expressions like 'game changer' and 'revolutionizing biology,' supported by references to validation scores, such as a GDT score of 92.4, and examples of how it accelerates understanding of proteins, highlighting both excitement and recognition of its profound influence.
Most discussions express strong support and excitement about AlphaFold's achievements, emphasizing it as a groundbreaking breakthrough in protein science.
The topics generated widespread curiosity, with many references to the scientific methods, applications, and potential impact on medicine and bioengineering.
The discussions include detailed references to scientific articles, validation results, and future implications, showing deep engagement with the material.
Many posts highlight the transformative potential of AlphaFold for research, drug discovery, and understanding life processes, indicating high perceived impact.
Abstract in brief
- Proteins are crucial for life, and knowing their structures helps us understand how they function, but only about 100,000 out of billions of known proteins have had their structures determined due to the lengthy experimental process.
- Developing accurate computational methods to predict protein structures has been a major challenge for over 50 years, especially when similar structures are not available.
- The new version of AlphaFold is a groundbreaking neural network model that achieves atomic-level accuracy in predicting protein structures, validated by performance in the CASP14 competition, and uses advanced machine learning techniques that incorporate biological principles.
Top posts
Most liked on X and BlueskyAlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」みたいのを開講したら1セメスター丸ごと使うか,足りないレベル https://t.co/MIjr9zmVKa
Speaking now: John Jumper @DeepMind "Highly accurate protein structure prediction with AlphaFold" https://t.co/QeUzoPkkkO
Makale: https://t.co/8ENGmAneGT Kodlar: https://t.co/n2u4I37lCX Yardımcı materyaller: https://t.co/vwmRkC7jNV
Here is my implementation of Invariant Point Attention from the "Highly accurate protein structure prediction with AlphaFold" paper It is a form of attention acting on a set of frames and is invariant under global Euclidean transformations on the frames https://t.co/QjhPL870MD
All mentions
Latest Scientific Breakthroughs Including AlphaFold and Recyclable Polymers
We present to you biggest research breakthroughs in Science particularly chemistry and drug discovery in the year 2021. The major one being the predictive ability of Alphafold, skeletal editing, MDMA trials, recyclable…
Significant Advances in Physics in 2021 Highlighted
AlphaFold: Breakthrough in Protein Structure Prediction
AlphaFold is a groundbreaking neural network model that predicts protein structures with atomic-level accuracy. This video explains how the program works, its capabilities, and limitations, highlighting its significance…
Advancements in Protein Structure Prediction with AlphaFold
Proteins are essential for life, and determining their structures helps us understand their functions. AlphaFold's neural network model achieves atomic-level accuracy in predicting protein structures using advanced…
Advances in Protein Structure Prediction Using AlphaFold's Deep Learning Techniques
Proteins are crucial for life, and understanding their structures helps us understand their functions. AlphaFold's neural network model achieves atomic-level accuracy in predicting protein structures, revolutionizing…
Alphafold2 - one of the most scientific breakthroughs of the decade - delivers state-of-the-art end toend protein structure prediction with near-experimental accuracy! papar - https://t.co/XjnG028HxG code - https://t.co/GXXwEBgvgy (1/8) #proteinchallenge #DeepLearning https://t.co/yCbSm8JCAl
11/
Refs cont. Jumper, John et al. “Highly accurate protein structure prediction with AlphaFold.” Nature vol. 596,7873 (2021): 583-589. doi:10.1038/s41586-021-03819-2
The role of metabolism in shaping enzyme structures over 400 million years Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). Article ADS CAS PubMed PubMed Central Google Scholar Opulente, D. A. et al. Genomic factors shape carbon and…
Olaf Ronneberger (of U-Net fame) was also on the AlphaFold paper: scholar.google.com/citations?us...
RT @LongevityPapers: 7/ Next paper in this series is the #Alphafold 2021 @Nature paper from @demishassabis group at @GoogleDeepMind led by…
@Ali82194556 According to a recent publication in Nature, 'AlphaFold achieves a median global distance test (GDT) score of 92. 4 across all targets, outperforming the next best method by 25. 8 GDT points' (Jumper et al. , 2021, doi: 10. 1038/s41586-021-03819-2).
Highly Accurate Protein Structure Prediction with AlphaFold [in English] https://t.co/kJKQtU6IoT via @YouTube
1/8
RT @LongevityPapers: 7/ Next paper in this series is the #Alphafold 2021 @Nature paper from @demishassabis group at @GoogleDeepMind led by…
RT @LongevityPapers: 7/ Next paper in this series is the #Alphafold 2021 @Nature paper from @demishassabis group at @GoogleDeepMind led by…
RT @LongevityPapers: 7/ Next paper in this series is the #Alphafold 2021 @Nature paper from @demishassabis group at @GoogleDeepMind led by…
7/ Next paper in this series is the #Alphafold 2021 @Nature paper from @demishassabis group at @GoogleDeepMind led by #John_Jumper Highly accurate protein structure prediction with AlphaFold https://t.co/6MHAtGuLQS Jumper and @demishassabis were awarded 2024 Nobel Prize in https://t.co/FddDvBzpBL
RT @Atsushi_twi: SciPyを使っているという証拠。Nature論文より。これがOSSの醍醐味だと思うな。あなたもノーベル賞に貢献的な。“Highly accurate protein structure prediction with AlphaFold”ht…
SciPyを使っているという証拠。Nature論文より。これがOSSの醍醐味だと思うな。あなたもノーベル賞に貢献的な。“Highly accurate protein structure prediction with AlphaFold”https://t.co/vLHJWAeIyN https://t.co/3qiZN1lmAk
RT @Kaochar2: Highly accurate protein structure prediction with AlphaFold | Nature https://t.co/PQ1CBkWF6D
RT @Kaochar2: Highly accurate protein structure prediction with AlphaFold | Nature https://t.co/PQ1CBkWF6D
Highly accurate protein structure prediction with AlphaFold | Nature https://t.co/PQ1CBkWF6D
RT @RSGTurkey: Herkese selamlar
Herkese selamlar
Highly Accurate Protein Structure Prediction with AlphaFold | SimonKohl https://t.co/XKqRHJExQB via @YouTube
AF2 https://t.co/iT141105cY AF2 doc https://t.co/BLiGoxn7jH
これか、すごい世の中になったものだ Highly accurate protein structure prediction with AlphaFold https://t.co/fnR8Hpz4TN
@AdamShubinsky @GillVerd @FutureJurvetson https://t.co/NL7yXDsqtt
@_nasch_ solo IF's: https://t.co/Knls6JatEc https://t.co/3C0tDxeNNc https://t.co/VHKrSzP9h2
Fun fact: today the #AlphaFold2 paper reached the amount of ~20,000 citations* in only 2 years and 9 months!
Highly Accurate Protein Structure Prediction with AlphaFold | SimonKohl https://t.co/O3t7KYhXxU 来自 @YouTube
Highly Accurate Protein Structure Prediction with AlphaFold | SimonKohl https://t.co/wHJB3m509T via @YouTube
RT @sugargroove: この論文にはCASP14でのAlphaFold2の予測精度はCα RMSD(95% residue coverage)のmedianで0.96 Åと書いてある。 しかし自分の未発表の構造と今のColabFoldの予測構造だとCα RMSDは0.…
RT @sugargroove: この論文にはCASP14でのAlphaFold2の予測精度はCα RMSD(95% residue coverage)のmedianで0.96 Åと書いてある。 しかし自分の未発表の構造と今のColabFoldの予測構造だとCα RMSDは0.…
5/ Alphafold2는 단백질 구조의 학습, 예측, 수정, 개선 등 구조 예측에 필요한 모든 단계를 수행하는 엔드투엔드 모델로, 더욱 자세한 기능에 관심 있으신 분들은 아래 논문 및 Supplementary information을 참조하시면 좋을 것 같습니다. https://t.co/ZHTiNbOQ07 https://t.co/hPVkh4FkEZ
この論文にはCASP14でのAlphaFold2の予測精度はCα RMSD(95% residue coverage)のmedianで0.96 Åと書いてある。 しかし自分の未発表の構造と今のColabFoldの予測構造だとCα RMSDは0.4-0.5 Åあたりになることが多く、1 Å近くになることは稀。 https://t.co/tlRcRPTqlh
In addition to Demis Hassabis and John Jumper, this was a team of 37 total with 20 co-lead authors (according to the paper). A lot of them have interesting scientific careers. pubmed.ncbi.nlm.nih.gov/34265844/
Highly accurate protein structure prediction with AlphaFold https://t.co/FmJOqlTRc5 We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Critical Assessment of protein Structure Prediction (CASP14) #AlphaFold (10/n) https://t.co/G3eAAfwNhH
RT @IndabaxMA2023:
RT @IndabaxMA2023:
RT @IndabaxMA2023:
RT @IndabaxMA2023:
RT @IndabaxMA2023:
RT @IndabaxMA2023:
RT @IndabaxMA2023:
time to study how the LHS (RosettaFold's Biased Axial Attention) bypasses the need for RHS (AlphaFold's triangle multiplication / inequality) https://t.co/4T3hyyWY2F https://t.co/CESEJdIWjW https://t.co/WW7Tsh8udT https://t.co/wIMACkZalc
@BornsteinMatt @karpathy @derrickharris @appenz For the AlphaFold paper, I'd recommend reading the other paper "Highly Accurate Protein Structure Prediction with AlphaFold" instead as it introduces transformers to the mix with their 'evoformer'. Imo this paper is far more pivotal to the field
@lyndonliskovec Much of it goes over my head, but feel free to look into the official documentation for AlphaFold at https://t.co/qKRe2gXlQm It’s also open source from what I’ve heard, so I’m sure there are videos of people using it (maybe The Thought Emporium had a Whose Gene episode with it?)
RT @Triangl3Agency: Highly accurate protein structure prediction with AlphaFold – La Biblia de la IA – The Bible of AI™ Journal https://t.c…
Highly accurate protein structure prediction with AlphaFold – La Biblia de la IA – The Bible of AI™ Journal https://t.co/8xocvN715Q #TheTriangleAgencyNews #accurate #AlphaFold #Automotive
RT @drifterddm: *Highly accurate protein structure prediction with #AlphaFold* by @twominutepapers Amazing work!!! https://t.co/035aYIoAal
*Highly accurate protein structure prediction with #AlphaFold* by @twominutepapers Amazing work!!! https://t.co/035aYIoAal
RT @0xthefool: 14 Highly Accurate Protein Structure Prediction With AlphaFold 作者:John Jumper等 时间:2021 DeepMind研究人员在AI应用的又一次开创研究,解决了约10万…
14 Highly Accurate Protein Structure Prediction With AlphaFold 作者:John Jumper等 时间:2021 DeepMind研究人员在AI应用的又一次开创研究,解决了约10万种独特蛋白质的三维结构问题,为科学家开发新药物和治疗疾病提供可能性。 https://t.co/DwqBwQS7I3
@GaryMarcus @primalpoly Are we thinking Anthromorphic Intelligence or Computational Intelligence? Using the computational power to solve science problems https://t.co/pcmSMYTIcx
RT @rishit_dagli: Here is my implementation of Invariant Point Attention from the "Highly accurate protein structure prediction with AlphaF…
Here is my implementation of Invariant Point Attention from the "Highly accurate protein structure prediction with AlphaFold" paper It is a form of attention acting on a set of frames and is invariant under global Euclidean transformations on the frames https://t.co/QjhPL870MD
RT @kanzure: AlphaFold2 (2021): "Highly accurate protein structure prediction with AlphaFold" https://t.co/QMpzMPeMkB https://t.co/47fwXsz…
AlphaFold2 (2021): "Highly accurate protein structure prediction with AlphaFold" https://t.co/QMpzMPeMkB https://t.co/47fwXszyUz
@AdrianoAguzzi Can't agree more. I don't want to do too much Google glorification but is the industry showing the exampe with this? 19 co-first authors on the AlphaFold paper https://t.co/BztXR3HZ1t https://t.co/0ZgOHK5X8T
Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873), 583–589. https://t.co/KCr36YYDwg Higgins, M. K. (2021). Can We AlphaFold Our Way Out of the Next Pandemic? Journal of Molecular Biology, 433(20), 167093. https://t.co/kMYmsF4a0f
MUST READ. Highly accurate protein structure prediction with AlphaFold (AI). See my evaluation at https://t.co/BgXSJzzcys. Original NATURE article at https://t.co/Sg98gJsmNa, PMID: 34265844. @deepmind @demishassabis https://t.co/jDkjtJQvoC
References to db used to mint MolNFT AF: https://t.co/duw3ekpX2k Jumper, J et al. Highly accurate protein structure prediction with AlphaFold. Nature (2021). https://t.co/VW32nfKpRv
Summer is over. New BIML Bibliography entry #MLsec (1/2) Highly accurate protein structure prediction with AlphaFold https://t.co/aaiH2RrgEZ A difficult to read paper (due mostly to unfamiliarity with the large number of subfields), but very interesting work. https://t.co/8zx7QYCGYf
RT @SpringerNature: You can read “Highly accurate protein structure prediction with AlphaFold” and 'Highly accurate protein structure predi…
You can read “Highly accurate protein structure prediction with AlphaFold” and 'Highly accurate protein structure prediction for the human proteome', published in @Nature, open access here: https://t.co/GsEQwyjJrN and here: https://t.co/4z24d1vvV1
Highly accurate protein structure prediction with AlphaFold https://t.co/fmEwfsDyC8
@davidbautistaqf háblales de: https://t.co/CTn5vc405r
RT @development_rt: RT @PinakiLaskar: RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #Mac…
RT @PinakiLaskar: RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashiontech #SmartCity #AutonomousVehicles #DataScience #BigData #Analytics #JavaScript #Serverless #Lin… https://t.co/aneptwsCbY
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashion…
Highly accurate protein structure prediction with AlphaFold #AI #IoT #IIoT #CX #MachineLearning #SelfDrivingCars #fashiontech #SmartCity #AutonomousVehicles #DataScience #BigData #Analytics #JavaScript #Serverless #Linux #RStats #Programming #Coding #NLP #Web3 #5G #Blockchain #ML https://t.co/4IJ9UpuJB4
we’ve also seen pretty good performance for transformers / attention based models in some physical tasks like protein folding, drug discovery etc. in part because of the more limited state spaces and some ability to simulate (e.g. alphafold) https://t.co/4o9aeCufAp
▶️Listen to PubReading [113] - Highly accurate protein structure prediction with AlphaFold - J. Jumper, P. Kohli, D. Hassabis et al https://t.co/oEvwK1SzpM
Highly accurate protein structure prediction with AlphaFold | Nature #news #Trending #AI #business #programming #Technology #DataScience #research @The_InnovationJ Tweets by @YangLiu_CTH https://t.co/25PQvAUAw2 https://t.co/YLsU0q4JbN
Highly accurate protein structure prediction with AlphaFold https://t.co/g2e7TDjpJ1 #AI #Research via @DeepMind
RT @future_of_AI: Highly accurate protein structure prediction with AlphaFold https://t.co/1w2Iv5JPc4 #AI #Business via @DeepMind
Highly accurate protein structure prediction with AlphaFold https://t.co/1w2Iv5JPc4 #AI #Business via @DeepMind
Highly accurate protein structure prediction with AlphaFold - Nature https://t.co/iAaqxcQuDd
Last talk before coffee: John Jumper from Google, "highly accurate protein structure prediction with AlphaFold." #bps2022
DeepMind: Highly accurate protein structure prediction with AlphaFold https://t.co/BqhbNK2jo0
@doux_sy recycleはAF2の論文(doi: 10.1038/s41586-021-03819-2)図1eのRecycling (three times)にあたります。
RT @CS_HSE: Watch @DeepMind's Anna Potapenko discussing highly accurate protein structure prediction with #AlphaFold at the latest Faculty…
Watch @DeepMind's Anna Potapenko discussing highly accurate protein structure prediction with #AlphaFold at the latest Faculty Colloquium: https://t.co/nki8JHWtUo
RT @States_AI_IA: Thank you very much for this work to @tfgg2, @mfigurnov and @Russb09 et al, via @States_AI_IA #AlphaFold #health #machin…
Highly accurate protein structure prediction with AlphaFold https://t.co/g2e7TDjpJ1 #AI #Research via @OriolVinyalsML
نظرة أخرى لمكتشفات العام 2021 العلمية https://t.co/dh6atfr8UC https://t.co/N7UrkhHIhU
RT @States_AI_IA: Thank you very much for this work to @tfgg2, @mfigurnov and @Russb09 et al, via @States_AI_IA #AlphaFold #health #machin…
Thank you very much for this work to @tfgg2, @mfigurnov and @Russb09 et al, via @States_AI_IA #AlphaFold #health #machinelearning #medicine #models #ai #artificialintelligence #thebibleai #openscience #openaccess #thanks https://t.co/LXwNUMGyu2
#2021年生命科学研究大賞 最優秀賞(独断・偏見) Highly accurate protein structure prediction with AlphaFold. やはり、Alphafold2。 今や改良により、プレプリレベルだとPPIや膜タンパク質の結晶構造予測も可能になるかも?とのこと。 https://t.co/5jgDnXDFn1
https://t.co/bzooskoylV Highly accurate protein structure prediction with AlphaFold
RT @workshopmlsb: Speaking now: John Jumper @DeepMind "Highly accurate protein structure prediction with AlphaFold" https://t.co/QeUzoPkkkO
Speaking now: John Jumper @DeepMind "Highly accurate protein structure prediction with AlphaFold" https://t.co/QeUzoPkkkO
RT @workshopmlsb: Featuring keynotes: John Jumper: “Highly accurate protein structure prediction with AlphaFold” Jane Richardson: “The earl…
RT @workshopmlsb: Featuring keynotes: John Jumper: “Highly accurate protein structure prediction with AlphaFold” Jane Richardson: “The earl…
Featuring keynotes: John Jumper: “Highly accurate protein structure prediction with AlphaFold” Jane Richardson: “The early days of structural biology before ML” And invited talks from @BarzilayRegina @mmbronstein @LucyColwell37 @CecClementi @keating_lab @Dereklowe w live Q&A! 2/
Сегодня, 7 декабря 2021 г., в 16:20 пройдет доклад: A. A. Potapenko, Highly accurate protein structure prediction with AlphaFold (https://t.co/b2KgYYOIxn). Коллоквиум Факультета компьютерных наук НИУ ВШЭ, г. Москва, https://t.co/HrLL9JM5VR
Highly accurate protein structure prediction with AlphaFold The explosion in available genomic sequencing technique & data has revolutionized bioinformatics but the intrinsic challenge of experimental structure determination did not keep pace - so far. https://t.co/CJzQ3yZSgB
https://t.co/9ucWLVkgrA 알파폴드 2 논문의 핵심은 Supplementary Information인데 정말 각잡고 논문썼다는 느낌이 팍팍
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
"Highly accurate protein structure prediction with AlphaFold"という、タンパク質の立体構造をAIで解明したという論文を読んで内容を発表した。論文の出版は2021年の7月(Natureにて)。すでに、500件近く引用されている画期的論文だ。 AIの世界はものすごい勢いで流れている。
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
RT @PinakiLaskar: Highly accurate protein structure prediction with AlphaFold #DataScientist #BigData #MachineLearning #Analytics #DataScie…
@kollmanlab @thesteinegger @sami_chaaban I think our excuse was that the protein is still pseudo-"folding" in the cartoon so it may generate physically unrealistic intermediates. As we see in animation here:
@uKitago こういうの (https://t.co/fQLyRb9bXo) を見るとタイミングによっては欲しい構造がたまたま出てくれたりしそうな気もしますね。もしくは、template に使われる構造を意図的に絞れたりすればより直接的に効くかもしれませんが。
Highly accurate protein structure prediction with AlphaFold https://t.co/R2BCZm2LXL
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
62 pages of suppl. Mat. Wonder what is really relevant in the main section. https://t.co/YEKN1gWLig
The latest version of #AlphaFold2, published in @Nature last month, now predicts protein structures with atomic accuracy, even when no homologous structures exist. (5/6) Link: https://t.co/biFU57mHvg
Regarding the exponential growth of knowledge... We've been "fully" acquainted with DNA for ± 15yrs, yet the broad structure of 100,000 proteins was determined using AI-Alphafold, a process that was going to take us years to accomplish! #medtwitter https://t.co/031FyxOSct
RT @EnrikeLopez: Portada histórica de @el_pais: "Predicción de alta precisión con AlphaFold de la estructura de las proteínas" de John Jump…
RT @EnrikeLopez: Portada histórica de @el_pais: "Predicción de alta precisión con AlphaFold de la estructura de las proteínas" de John Jump…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
This could be a game changer! https://t.co/pylM4kaCiz
https://t.co/w7vGc0Tfaz https://t.co/2OaCUxCVBT El Dr. Andrei Lupas, biólogo evolutivo del Instituto Max Planck de Biología del Desarrollo y juez del CASP: “Esto cambiará la medicina. Cambiará la investigación. Cambiará la bioingeniería. Lo cambiará todo" https://t.co/jOHH8Ly9kD
Portada histórica de @el_pais: "Predicción de alta precisión con AlphaFold de la estructura de las proteínas" de John Jumper et. al. Nature, 2021 https://t.co/w7vGc0Tfaz https://t.co/3yMzowy411 "...facilita comprender el funcionamiento de las #proteínas, esenciales para la vida" https://t.co/Ymr4oJhnzi
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @LindorffLarsen: @Alfons_Valencia @sokrypton @minkbaek Certainly tempting, but I'm pretty sure it isn't in general. Both based on readin…
Trending on PubMed: Highly accurate protein structure prediction with AlphaFold. https://t.co/oOLZdNNac9
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @syoifczeri: https://t.co/gXHCJs9SBA つまり、ただのTransformerではなくここのAlgorithm6のEvoformerみればコンタクトマップあたりの予測できるんやな?(普通のTransformerつかって過学習に陥った人並み…
https://t.co/gXHCJs9SBA つまり、ただのTransformerではなくここのAlgorithm6のEvoformerみればコンタクトマップあたりの予測できるんやな?(普通のTransformerつかって過学習に陥った人並みの感想
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
AlphaFold: "Highly accurate protein structure prediction with AlphaFold" Jumper J, Evans R [..] Hassabis D. Nature. 2021-07-15 https://t.co/CSnmfdQjm8 https://t.co/XCtIbCXQbh
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
RT @ImAI_Eruel: AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」…
AlphaFold2については,論文自体は手法解説の本体ではなく,同ページの「Supplementary information」が、実質的に論文の本体なのですが,これがまた凄まじいボリュームで,大学の専門の授業で「AlphaFold2完全理解」みたいのを開講したら1セメスター丸ごと使うか,足りないレベル https://t.co/MIjr9zmVKa
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @oktayarslan: Makale: https://t.co/8ENGmAneGT Kodlar: https://t.co/n2u4I37lCX Yardımcı materyaller: https://t.co/vwmRkC7jNV
Makale: https://t.co/8ENGmAneGT Kodlar: https://t.co/n2u4I37lCX Yardımcı materyaller: https://t.co/vwmRkC7jNV
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
RT @MWNautilus: Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi…
Finally, the long awaited #AlphaFold 2 methods for #3D #protein #structure #prediction were published: https://t.co/O9CTWqi0Ru ...along with the #OpenSource code on @github: https://t.co/qXqlC0E53h #DIYBio #AI #ML #DeepMind #biohacking #KI #TensorFlow #bioinformatics #OpenPharma https://t.co/PrvgkkCQWs
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @souyakuchan: #AlphaFold 2 論文 Supplementary Video 2。 2180 残基と大きい crAss-like phage RNA polymerase (CASP14 出題: T1044) の構造予測。 縮こまった状態(?)から正…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
RT @LindorffLarsen: Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.…
Highly accurate protein structure prediction with AlphaFold https://t.co/ctBzyWnaHn https://t.co/45WfmPHwHL https://t.co/IYexnQd1oj
Highly accurate protein structure prediction with AlphaFold https://t.co/bWhDsldyQv
C
I
勤万