Pranam Chatterjee
@pranamanam
Followers
3K
Following
1K
Media
44
Statuses
541
Designing proteins to program biology! ๐งฌ๐ป๐งซ Assistant Professor at @DukeU | Co-Founder @GametoGen and @UbiquiTxINC | Formerly @MIT '16 '18 '20 and @harvardmed
Cambridge, MA
Joined August 2015
I'm going to keep saying this until someone proves me otherwise: AlphaFold(2, 3, 10000, whatever) or any structure prediction tool will not generalize to protein conformations if its training is optimized to "ground truth" structures in the PDB! Crystallized or cryo-EM structures
27
59
364
I'm usually not too emotional about paper acceptances, but this one justifies it. ๐ฅน It doesn't feel real, but PepPrCLIP is now published at @ScienceAdvances! Let me tell you its story. ๐. ๐: ๐ป: PepPrCLIP (๐ถ๏ธ๐) began as
11
55
376
Current protein models (ESM-2, AlpaFold2,. ) only encode the 20 wild-type amino acids -- what about PTMs, which significantly influence the diversity of the proteome? ๐โโ๏ธTo solve this, we present the first PTM-aware protein language model, PTM-Mamba!
7
72
348
Had a day to reflect on the release of ESM3, and just wanted to share a few thoughts (and a few shameless highlights of my lab's work! ๐
). Before that, for the people who know our stuff, you know that I am an ESM evangelist: I think pLMs will be the future of protein design. ๐ช
6
65
320
A beautiful thread from @rohitsingh8080!! ๐งต It really captures my feelings/frustration about current "trends" in protein design. As Rohit mentioned, the aftermath of sequencing the human genome has been a huge huge disappointment. Genomics has really not lived up to the hype,.
Let me tell you a story. It'll end up at the current tech-bio and protein design scene. But the story starts about 25 years earlier. Did you know that, commercially, the human genome project precipitated the end and not the start of a genomics boom? 1/.
17
38
296
AlphaFold3 is out -- and it's all-atom, diffusion-based, with no equivariance constraints! ๐งฌ For biologists, now you can model PTMs, ions, DNA/RNA, small molecules -- but only 10 jobs/day. Unfortunately for us in the TechBio community, no code (though a lot of replication
8
54
248
What if you could design binders to specific spots/motifs on target proteins: conserved epitopes on pandemic viral phosphoproteins, disordered regions on dysregulated transcription factors, or even breakpoints on fusion oncoproteins? That would unlock unprecedented specificity
2
47
236
Seems like a hot day for peptide models! ๐ฅ To this point, we've been modeling peptides as linear amino acid sequences via ESM-2 within our binder design models, SaLT&PepPr, PepMLM, and PepPrCLIP. However, most therapeutic peptides are chemically modified, enhancing their
4
42
207
Designing proteins is all fun and games, and yes, I believe sequence-based models do it better. ๐
BUT, at the end of the day, it only matters if we can actually have a meaningful impact with the proteins we design. And so I am SUPER EXCITED to show our new collaborative data on
7
30
188
Super excited to share our new protein language model: FusOn-pLM, developed by my wonderful first-year PhD student @SophieVincoff and her team of amazing undergrads!! ๐ . Sure, we know AlphaFold3/RFDiffusion and other structure prediction/design methods are great -- but what
8
38
178
Hi everyone! Excited to share that I will join the amazing faculty at @DukeU as an Assistant Professor of Biomedical Engineering in July! We will design therapeutic proteins for genome editing, proteome editing, and ovarian cell engineering! ๐ปโ๐งฌโ๐งซโ๐ฅ.
24
5
157
Really interesting paper: ESM-all atom (ESM-AA) to compete with structure-focused Rosetta-AA and AlphaFold3! โจ ESM-AA uses multi-scale code-switching to model molecules at residue and atom scales, using multi-scale position encoding to capture contexts in both regimes. It seems
2
30
155
With the hype around ESM3, I just want to show how impactful pLMs can be, even beyond biomedicine! ๐ I'm really excited to present MetaLATTE, our brand new metal binding predictor trained on ESM-2 latent embeddings via a super unique multi-task learning strategy, now live on
3
40
142
#NeurIPS2023 was so much fun! My favorite talk from the workshop has to be from @PrescientDesign! Similar to @OpenAI's Consistency Model formalism, they sample a score-based data manifold with one step denoising for antibody design! #GenerativeAI.
3
21
135
This is the beauty of collaborative science! ๐งช Over the summer, my good friend @htkratochvil and I were thinking about how we could generate membrane proteins (her lab's expertise) with our lab's generative pLMs. Obviously, structure-based methods aren't great as membrane
4
22
132
An exciting @biorxivpreprint update on our recent PepPrCLIP model!๐ถ๏ธ๐ Hopefully, you all remember PepPrCLIP, where we apply Gaussian perturbations to the peptidic latent space of ESM-2 to de novo generate naturalistic peptides, and then input these new peptide sequences into a
0
21
128
Been working on a lot of our lab's manuscripts lately (stay-tuned), but just wanted to share one of my favorite papers I've been reading: microenvironment-aware hierarchical prompt learning to predict ฮฮG changes in PPIs upon mutation! ๐คฉ. We've seen a lot of ฮฮG prediction
1
27
126
Interesting new pLM (AA-0) from @Ginkgo! It definitely seems they have found, as we've discovered many times over, that ESM-2-650M is the ideal baseline pLM, but Ginkgo does make some notable "enhancements" with their model. Architecturally, nothing different, but the training
3
14
124
This is my favorite cell embedding paper! SATURN learns universal cell embeddings by coupling gene expression with pLM embeddings for cross-species integration of functionally-related genes (macrogenes). ๐ Amazing for atlas integration!.Paper: Code:
3
23
116
Okay, this is SO COOL from @Mila_Quebec: Aaren, which leverages a parallel prefix scan algorithm, can be trained in parallel (Transformers) but only requires constant memory (RNNs)! ๐คฉ It's kind of similar to RWKV and Linear Attention, but unlike those methods, Aaren exactly
0
20
110
Our new preprint is up -- and it's completely de novo! ๐คฉOur Peptide Prioritization via CLIP (PepPrCLIP) model is the first de novo binder design algorithm only needing the target amino acid sequence (no 3D structure needed). ๐ถ๏ธ๐Take a read here!
4
21
108
Excited to share our newest Cas9 out in @NatureComms! ๐ฅณ We recombine our Sc++ enzyme (NNG) with the PID of @BKleinstiver's SpRY (NRN) to generate a PAM-flexible chimeric Cas9 = SpRYc! ๐ถ๏ธ SpRYc can edit at diverse NNN loci -- take it out for a spin! ๐งซ
3
27
106
We've updated PepMLM with new comparisons to RFDiffusion! ๐Without using structure, PepMLM has a higher in silico hit-rate for peptide generation on structured targets (those with existing peptide binders) than RFDiffusion. More in vitro data coming soon!
3
17
99
Exciting news from the lab! ๐ We have updated PepMLM on the @arxiv with new computational and experimental benchmarking results! ๐ฅณ . Just as a reminder: we trained PepMLM via a novel span masking strategy that positions peptide binder sequences at the C-terminus of their target
5
21
98
Whoa!! @EvoscaleAI just came out with a potentially useful upgrade to our favorite ESM-2 pLM with three new ESM C representation models: 300M, 600M, and 6B! ๐ณ Here, I'll do three semi-analyses based on the release: model modifications, "evaluations", and license. Would be happy.
Introducing ESM Cambrian. Unsupervised learning can invert biology at scale to reveal the hidden structure of the natural world. Weโve scaled up compute and data to train a new generation of protein language models. ESM C defines a new state of the art for protein
3
9
94
๐งต (1/4) As a brand-new PI, I'm psyched to present my group's first preprint! Here, we apply @OpenAI's CLIP architecture to design target-specific peptides, fuse them to E3 ubiquitin ligases, and degrade pathogenic proteins in a pipeline called Cut&CLIP. โ๏ธ
7
11
90
Diffusion has worked incredibly well for direct protein structure (RFDiffusion) and sequence (EvoDiff) generation! ๐ Our new AMP-Diffusion model applies latent diffusion to ESM-2 pLM embeddings to generate diverse, naturalistic AMPs! Check it out:
1
19
88
Just getting around to posting about this beautiful paper from @DreamFoldAI!! Learning a joint representation based on SOTA sequence/structure embeddings and decoding the joint representations of the inputs into SE(3) vector fields is very clever -- my hunch is that the ESM-2
We are thrilled to announce FoldFlow-2: our new SOTA protein structure generative model. w/ @Guillaume_hu James V @FatrasKilian Eric TL @PabloLemosP @riashatislam @ChengHaoLiu1 @jarridrb @tara_aksa @mmbronstein @AlexanderTong7 @bose_joey. ๐ช Blog post:
1
18
90
Surreal! ๐คฉ Along with @martinvars and @DinaRadenkovic, we co-founded @GametoGen in 2020 with just a silly graph theory algorithm I designed to predict TFs that could differentiate ovarian cells from iPSCs. ๐ป Now, little Mia is here with the technology that has grown out of that.
(1/) We are thrilled to announce a medical milestoneโthe worldโs first live birth using our product, Fertilo, that matures eggs outside the body.
8
7
84
I am 100% on the side of the DiffDock team on this one, even as a someone who believes almost all structure-based docking models are artifactual. Some things to note:. -The more "standard" docking pipelines that were compared against DiffDock on head-to-head tasks are largely.
You may have seen a recent pre-print [1] from Jain et al. with strongly worded claims against the experimental results in our DiffDock paper [2]. We initially declined to respond as we saw that this preprint contained falsehoods, misleading comparisons, seemingly deliberate.
5
8
79
Today, I "officially" begin my faculty position at @DukeU! ๐ฅณ Super excited to welcome my wonderful team of postdocs, PhD/Masters students, and undergrads from around the globe to our little but mighty lab in Durham โ let's change the world together! ๐ช๐พ
5
7
76
SaLT&PepPr is published in @CommsBio! ๐ฅณ Here, we fine-tune the ESM-2 pLM to identify peptidic binding sites on target-interacting partner sequences. We fuse these "guide" peptides to E3 ubiquitin ligases to degrade disease-causing proteins! ๐ปโก๏ธ๐งซ (1/n)
5
15
74
A great day for our field, and very well deserved!!! ๐
The work done at @UWproteindesign and @GoogleDeepMind, and of course that of @geoffreyhinton @HopfieldJohn, and the RNAi/miRNA community, especially at @UMassChan, have surely changed my life, and have inspired so many.
BREAKING NEWS.The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Chemistry with one half to David Baker โfor computational protein designโ and the other half jointly to Demis Hassabis and John M. Jumper โfor protein structure prediction.โ
1
9
73
Two really cool papers to read this weekend: one new dataset and one new algorithm! ๐ง. First, an incredible new dataset of protein-coding variations from exome sequencing of 983,578 diverse individuals -- over a million new variants! ๐คฉRetrain ESM-1v anyone? .Paper:
1
7
71
How can AI help us cure rare diseases? In the Aug/Sep @WorkingAtDuke magazine, I talk about my lab @DukeUBME and how weโre training generative language models to help us design drugs to target complex proteins. Keep an eye out for the magazine in your mailbox! @dukeresearch
1
8
72
Such a beautiful study from our friends in @rohitsingh8080โs lab here at @DukeU! An autoencoder on top of ESM-2 for fixed dimensionality is brilliant, and perfectly suited for this problem, one that has plagued delivery and other drug design applications. ๐ I can imagine so many.
De novo protein design is great, but nature has millions of proteins- why not repurpose them? . Introducing Raygun, a new approach to protein design. It allows you to miniaturize, magnify or modify any protein. We synthesized miniaturized variants of eGFP and mCherry! 1/
2
16
70
Unique application of GPT-like LLMs to RNA! Model-prioritized mutations (via relative log likelihood to WT) seem to do pretty well on improving thermostability of 23S rRNA sequences. Always exciting to see sequence-based models do well for design. ๐
0
14
68
So many congratulations on to my very brilliant friend @MartinPacesa on this really amazing achievement with BindCraft!! It's a very "crafty" use of AF2-Multimer weights and super principled validation and filtering with AF2 metrics and Rosetta physics-based scoring. metrics. ๐.
Have you ever wanted to design protein binders with ease? Today we present ๐ฉ๐๐๐
๐ช๐๐๐๐, a user-friendly and open-source pipeline that allows to anyone to create protein binders de novo with high experimental success rates. @befcorreia @sokrypton.
1
8
68
Try out @pengzhangzhi1's PyTorch implementation of ESM-2 with FlashAttention! ๐ฆ So easy to use that I even I ๐ต was able to drop it into some of my own code and get it to work in seconds. It's always so nice to have Fred making life easier for us in the lab!! ๐ซถ.
ESM2 is our go-to pLM. However, running, fine-tuning, or training a model with millions of parameters can be costly. We reimplemented ESM2 with FlashAttention, achieving up to 60% memory savings and 70% faster inference! ๐Check it out on github: (1/n)
1
7
66
Hey guys, I think I may have shared too much about my feelings (as usual) about protein design in the last post! ๐
I definitely don't mean to put down any line of research, and I think all scientific exploration is important -- that's why I love being a scientist and engineer!.
Interested in how #LLMs are used in #health? I'm so excited to have @pranamanam giving the first virtual seminar of the semester in the @DukeEngineering Computational Health Seminar Series. Anyone can join in remotely to learn more about his innovative approach!
7
2
64
Okay, another VERY exciting paper drop! ๐๐ So you guys may know the story of how @GametoGen was started? ๐ฃ My part of the story begins with a visit to @geochurch at @harvardmed, with a (perhaps naive) goal: to make an oocyte from a stem cell. ๐งซ With @GametoGen's support, I
0
11
65
Hi everyone! I am very excited to announce that our broad, efficient, and specific enzyme, Sc++, has been published in @NatureBiotech! ๐คฉ An incredible collaboration with @njakimo @JooyoungLee10 @SontheimerLab @medialab @UMassMedRTI ๐๐ Check it out! ๐ฝ (1/5).
An engineered ScCas9 with broad PAM range and high specificity and activity
3
15
60
This is such cool work published in @NaturePhysics! ๐คฉ An N-terminal unstructured and flexible glycineโserine peptide tag can accelerate nuclear import rates for stiff protein cargos by increasing their deformability. Definitely useful for my experimentalists to deliver our
0
4
60
Well here it is: ESM3!! ๐ The training regime is a bit different than your classic ESM-2/BERT model: you can start with a fully masked sequence and iteratively unmask. Super similar to our PepMLM model with ESM-2, which showed significant promise at full masking and unmasking.
We have trained ESM3 and we're excited to introduce EvolutionaryScale. ESM3 is a generative language model for programming biology. In experiments, we found ESM3 can simulate 500M years of evolution to generate new fluorescent proteins. Read more:
1
9
56
Our first deep dive into reproductive bioengineering! Here, we identify two transcription factors that, when overexpressed in iPSCs, generate functional granulosa cells, which support germ cell maturation, follicle formation, and steroidogenesis! ๐คฉ
3
5
52
Such a clever new LM architecture: Orchid uses data-dependent convolutions to overcome the quadratic complexity of traditional attention. ๐บ Achieves quasilinear scalability with dynamic kernel adjustments for long sequences! Also, super strong as a BERT model! . Paper:
2
11
50
I'm so excited to co-host the @gembioworkshop at #ICLR2024 in Vienna!! ๐ฅณTo both my computational and experimental friends, please reach out if you are interested in submitting a paper or serving as a reviewer -- it should be an incredible, one-of-a-kind meeting! ๐.
Announcing the Generative and Experimental Perspectives for Biomolecular Design workshop at #iclr2024!.We hope to bring together researchers in ML and experimental biology to accelerate progress on real-world applications. Website: .Paper deadline: Feb 3
3
11
51
KANs seem really cool! ๐Stemming from the Kolmogorov-Arnold representation theorem, they introduce learnable activation functions on edges, rather than traditional fixed activations on nodes! Still need to figure out which activation functions are good -- not too clear from the
2
11
52
You guys know how much of a sequence-first mindset I have! ๐
Based on experimental evidence from my own research/lab and our papers, I really do believe that pLM-based generative algorithms can have an impact where structure-based generators fail (which for proteins we work on,.
๐จDiscrete diffusion models are the next big thing. But guess what? RLHF-style fine-tuning and guidance for them arenโt the answer! โ. What does work? ๐The simulation-free DDPP objective. Check out our latest work for all the details. ๐ a ๐งต. Arxiv:
0
9
52
A little late, but super excited to share a great collaborative work by our team at Duke, my postdoctoral lab of @geochurch, and my company @GametoGen to engineer a functioning human ovarian follicle from iPSCs using our STAMPScreen method! ๐งซ
1
7
49
After almost three years of painstaking effort, today, I'm excited to present my team's amazing work on specifying human oogonia via combinatorial transcription factor induction in iPSCs! ๐ฅณ An incredible collaboration with my good friend @krammecc!
2
4
41
Beautiful stuff @AlexanderTong7 @mmbronstein and team!! MFM enables generative models to accurately match vector fields on data manifolds -- so many cool applications: LiDAR navigation, unpaired image translation, single-cell RNA trajectory inference. such impressive work! ๐.
If data lives on a manifold, how do we design meaningful interpolations between marginals? We present Metric Flow Matching (MFM)โฆ. @PPotaptchik @TeoReu @leoeleoleo1 @AlexanderTong7 @mmbronstein @bose_joey @Francesco_dgv . ๐Dive in here: ๐งต (1/12)
1
11
43
So excited to host the 2nd GEM Workshop at ICLR 2025! ๐ We have amazing speakers/panelists ๐งโ๐ฌ, money for new AI+Experiment collabs ๐ค๐ค, and we're partnering with @NatureBiotech to get the best papers into review! ๐ Definitely submit your new work and see you in Singapore!! ๐ธ๐ฌ.
We're BACK! ๐งฌ The 2nd GEM Workshop at @iclr_conf 2025 is ON!! ๐ Exciting speakers, exp-comp matchmaking + seed grants, and a partnership with @NatureBiotech! ๐ Submit your paper and join us in Singapore! ๐ธ๐ฌ. Website: Papers Due: February 3rd, 2025 ๐.
0
5
43
๐จ Current graduate students! If you're interested in developing and leveraging generative language models for therapeutics design, please apply to the @FutureHouseSF postdoctoral fellowship and indicate my lab as an option! ๐ $125k salary and access to all of @FutureHouseSF's.
FutureHouse is launching an independent postdoctoral fellowship program for exceptional researchers who want to apply our automated science tools to specific problems in biology and biochemistry, in collaboration with world-leading academic labs. --$125,000 annual stipend.
0
7
42
Excited to host the ByteDance AI4Science for a seminar on Protenix! ๐ค Theyโre absolutely one of my favorite AI teams in industry and have done some beautiful, open-source engineering for the AIxBio community! ๐.
Interested in Protenix, the open reproduction of AF3 from Bytedance AI4Science team? Curious to learn more? Luck for you! Join us for a special talk hosted by Chatterjee Lab ( on Thursday, January 16, at 10 AM EST. Zoom: (1/n).
0
4
41
Excited that my first company @GametoGen has just raised our Series B!!๐งซ This is the amazing work of our incredible CEO @DinaRadenkovic and CSO @krammecc who have led the company to new heights since @DinaRadenkovic, @martinvars, @krammecc, and I decided to get this off the.
The AlleyWatch Startup Daily Funding Report for 5/29 ft:. STARTUPS: @GametoGen @authcaptives .FOUNDERS: @dinaradenkovic @martinvars @pranamanam @authcaptives .LEAD INVESTORS: @twosigmavc @firstmarkcap . . #NYCtech #startup #funding #VC.
4
3
41
A very cool use of graph-based diffusion for enzyme evolution! Also love the application to pAgo, which can be a very powerful programmable gene editing tool with improvements. #GenerativeAI #ProteinDesign.
0
9
39
Training a VAE-based representation model, performing diffusion on the latents to generate new aptamers, with SPR results show an average 3-fold Kd reduction? Probably needs a bit of refinement and validation, but so cool! ๐ซจ #GenBio.
0
4
38
Happy to share our early work on generating binding peptides conditioned ONLY on the target sequence! ๐PepMLM masks cognate peptides at the end of target protein sequences, and tasks ESM-2 to fully reconstruct the binder region. ๐ท #GenerativeAI
1
4
37
Very interesting study that shows AlphaFold3 captures a relatively global effect of mutations on PPIs by learning a smoother energy landscape, but doesn't seem to be as atomicallly fine-grained as standard MD. Could still be good for generating synthetic mutant datasets
2
2
39
Hi all! We have released a preprint on our novel SaLT&PepPr language model to design peptide-guided degraders using PPI information. ๐ฅณ An incredible effort from @garykbrixi and team at @DukeU, and our collaborators at @Cornell with @mattdelisa! ๐ค๐พ๐ค๐พ
1
6
34
Today is my happiest day as a scientist: my incredible undergrad @SabrinaKoseki was awarded the @NSF GRFP! ๐พ I've mentored Sabrina for 3+ years now, and her brilliance, curiosity, and diligence EPITOMIZES scientific excellence. I've been so lucky to have her as my student!๐
2
1
33
So I know this is Christmas Eve๐, but I am so excited to share this INCREDIBLE breakthrough algorithm from my lab! I promise I'm not exaggerating: let me explain why! ๐. I'm sure you guys all know about Ozempic and Wegovy by now? ๐
GLP-1R agonist peptides, which when
3
5
34
@NeurIPSConf Not surprised that the Spotlight paper is from a school like Groton. Letโs stop putting random pressures on high schoolers โ this is really really getting out of hand. ๐ In a few years, a NeurIPS acceptance will be almost mandatory to get into places like MIT/Caltech/etc.
0
0
32
The Programmable Biology Lab is all set up at @DukeU โ all in 1.5 days! ๐ช๐ฝ Shoutout to my amazing Harvard CS undergrads, Garyk and Suhaas, for helping make this happen. ๐Time to design some programmable proteins! ๐ปโก๏ธ๐งซ
5
0
31
Nice! A great application of Mamba on concatenated, homologous protein sequences! ๐ Happy the authors mentioned our recent PTM-Mamba model, where we showed the application of Mamba to represent modified sequences! Overall, SSMs seem to have a found nice home in biology! ๐งฌ.
๐ Excited to introduce ProtMamba! Our novel protein language model designed to facilitate protein design.๐ฌ๐ ๐ง 1/n
2
5
31
Super excited to announce the publication of our "AA" targeting Cas9, iSpyMac, in @NatureComms! ๐ Another incredible collaboration with @njakimo @JooyoungLee10 @SontheimerLab @medialab @UMassMedRTI ๐Check it out ๐ฝ (1/5).
5
5
29
Check out our lab's perspective on generative design of therapeutic binders published in COBME: ๐ป๐งซ Great work from my PhD students:.@LeoChan213 @lauren_hong11 @vivi_64_ @SophieVincoff Feel free to cite the review in your next manuscript! ๐#GenerativeAI.
1
9
30
My favorite paper of the week: CONDITIONAL preference optimization via mDPO to ensure that both visual and textual modalities are equally considered! ๐คฉArchitecturally, the authors introduce a reward anchor to maintain the likelihood of preferred responses, significantly reducing
0
9
31
We're so grateful to #EndAxD for believing in our binder-guided degrader technology, powered by #GenAI, to study and treat Alexander disease. ๐ป๐งซ Thank you to @thomas_wagner and little Max for inspiring my lab to study, treat, and cure AxD! ๐๐พ
5
3
28
So simple and clever: for new sequences, examine the loss patterns of tokens on a pre-trained model then train the new language model with a focused loss on tokens with higher excess loss. Would be fun to fine-tune pLMs with this method! .Paper:
1
3
27
Remember @DeepMind's protein folding algorithm, AlphaFold2? I'm excited to share our team's early work on our own protein structure prediction pipeline! (1/3)
1
7
24
As the second week of my lab's existence comes to a close, I'm super grateful for the dedication and commitment of my students to our (quite lofty) goals. Yeah, we may not be the most well-known bunch, but we're about to do big things! ๐ช๐พ #FearlessFriday.
1
0
23
My last day together at the @medialab with my superhero team of undergrads, @SabrinaKoseki, Teodora, and Emma! So very proud to have watched you three become such brilliant young scientists, and so privileged to have mentored you as my first students! ๐๐ฅน
0
0
24
An integrated pLM trained on BOTH sequence and structure! ๐คฉ We're excited to leverage these embeddings in our lab -- the T5 architecture is super powerful for design! So many congrats to.@HeinzingerM and the @rostlab for another pioneering work! ๐
1
2
23
Residue energy-based preference optimization of antigen-specific antibodies. ๐ซจ A very interesting way to get around the lack of diverse paired structures for training!.
How to steer generative antibody design models to design antibodies satisfying user-desired preferences, such as rationality and functionality๐ค? .We are thrilled to introduce AbDPO, a general framework for antibody design via energy-based preference optimization. The preprint,
0
1
23
Hi friends! Just a gentle reminder that ourย #ICLR2024 @gembioworkshop paper submission deadline is coming up in ONE week: February 3rd! We have multiple tracks that should be of interest to both the ML community and experimentalists!
0
5
22
Thereโs really no better than feeling than when your first mentees get the @NSF GRFP! @garykbrixi, youโre going to continue to be a scientific rockstar in grad school! ๐ You and @SabrinaKoseki are the most deserving recipients! Iโm so proud of you guys. ๐.
1
0
22
Very cool work leveraging library-on-library screening data of binding motifs with an ESM-based binding predictor. ๐ฅ pLM โ Experiments always wins! .
0
5
22
Hi all, I'll be presenting our team's work on designing target-binding peptides with generative language models (and other cool protein design work!) tomorrow @valence_ai! ๐ป๐งซ Please find Zoom details here: Looking forward to seeing you there! ๐ฅณ.
1
2
20
Beautiful beautiful work! Iโve said that pLDDT really tells you about how disordered proteins are โ now we can leverage AF2 with simulations to explore it! ๐ฅฐ Would be excited to see this extend to AF2-Multimer as well. ๐
.
Happy to share the publication of our paper in @Nature. Conformational ensembles of the human intrinsically disordered proteome. Work led by @GiulioTesei and @AnnaIdaTrolle. I'll post more later, but for now here is a link:.and a short movie about the work
0
1
20
Really excited to try this out! Btw, we should make it a norm for protein language models to be hosted on HuggingFace! ๐ค.
๐ฆ
the FAbCon has landed ๐ฆ
. Our #generative #antibody #LLM available on ๐ฃ FAbCon-small (144M).๐ฅ FAbCon-medium (297M) .๐ฆ
FAbCon-large (2.4B). Team effort @alchemabtx @all_your_bayes @aretasgasp @jhrf @jakegalson @JorgeNDias ๐.
1
3
20
Very excited to have our tour-de-force genetic screening paper out in @CellRepMethods! We've developed a powerful, integrated pipeline for identifying, screening, and interrogating genetic perturbations for defined phenotypic outputs. Let's review how STAMPScreen works! [1/5].
When scientists at the Wyss, @harvardmed, & @medialab got frustrated w/ existing tools & methods for gene engineering, they made their own. @PranamMIT, @krammecc, & colleagues invented STAMPScreen to help scientists get from a database to results fast.
3
5
18
@andrewwhite01 Miniproteins are this weird intermediate between the specificity of an antibody and the permeability/drug-like properties of peptides. I personally think miniproteins are a thing because diffusion/flow-matching are not ideal for antibodies or peptides. At least with the de novo.
0
1
17
An incredible paper that I have been excitedly anticipating is out in @Nature today -- so many congratulations to my brilliant friend and collaborator @MartinPacesa and the @MartinJinek lab on this groundbreaking structural elucidation of Cas9 conformational activation! ๐๐๐.
Very excited to see our work on Cas9 conformational activation out in @Nature ! What started as a very poor crystal structure back in 2017 in the @MartinJinek lab, turned into a full blown cryoEM project during the pandemic.
0
0
14
@Pandeylab @NatureComms Do you understand how dangerous it is to draw strong conclusions from poorly analyzed data with severe methodological flaws? People look at journals such as @NatureComms for legitimate conclusions that can affect hiring, public practice, etc. This paper should be retracted!.
1
0
14
A very cool new pLM from WWC-winning ๐ช๐ธ! Only 14.8 M trainable parameters via training on UniRef50, focusing primarily on enzymes. ๐ฎโ๐จ Generated seqs have plausible biochemical properties, but wish there was more comparative benchmarking with bigger pLMs.
0
3
13
Hi all! I hope you can take a few minute to learn about our latest research on designing target-binding peptides (amongst other useful proteins) using #GenerativeAI! Thanks so much to @valence_ai for having me! ๐๐พ
0
2
15
Read more about what my company @GametoGen is up to: It's crazy to see how far it's come since @martinvars and I founded it back in 2020! ๐คฏ @DinaRadenkovic and @krammecc continue to lead it to newer heights -- you guys are amazing! ๐.
1
0
14
Check out our new COVID-19 diagnostic platform, out today in @ScienceAdvances! Peptide beacons + mini-TIRF technology = sensitive S-RBD protein detection at femtomolar resolution. Really a great extension of our ubiquibody design work! ๐คฉ
2
2
14
@martinvars @DinaRadenkovic @GametoGen An wonderful story by @Forbes of @DinaRadenkovic's vision and leadership for @GametoGen that led us here: ๐.
0
1
13
Super excited about trying out CellREADR from our friends in Josh Huang's lab here at @DukeNeuro! This will be very useful for dynamic cell state engineering applications. ๐งซ
0
0
12
Hi everyone! We are hosting Duke AI Day (sponsored by.@nvidia) on our beautiful @DukeEngineering campus on June 7th!โ๏ธIf you work on developing or applying AI algorithms, we'd love for you to register (it's free to everyone!) and submit an abstract:
2
6
13
We are so grateful to #EndAxD for funding our research leveraging generative language models to design peptide-guided degraders of dysregulated GFAP! ๐Continued research support will allow us to develop this into a potent and specific therapeutic for Alexander disease! Please
0
1
12
So excited to have @sacdallago and @AlexanderTong7 join us @DukeU!! ๐ We're building such an amazing AIxBio community with @rohitsingh8080 @romerolab1 .and others! ๐ ESPECIALLY in all things protein/bio language models! ๐ป ๐งฌ Come join us in Durham! ๐.
Two major life updates: .- moving to Senior Applied Research Scientist in Digital Biology at NVIDIA (Jan '25).- starting a new lab at Duke as Visiting Prof (early '25) . Both roles focus on tackling hard problems in biological machine learning through collaborative research.
1
1
12
Check out our early-stage work on designing computationally-optimized peptides that bind to SARS-CoV-2 and recruit E3 Ubiquitin Ligases for degradation! Could be a potential alternative to PAC-MAN? #COVID19.
2
2
11
Okay, I know #AutoGPT is the AI hype right now, but check out the Regression Transformer! Discontinuous-masking-based autoregression for conditional protein sequence generation?! Yeah, my lab's definitely using this! ๐
1
2
12
Absolutely beautiful work from my talented and amazing friend, @JulesGrunewald!! If you're crazy about base editing, what's better than doing both C->T and A->G at the same time?! ๐งฌ๐พ.
A dual-deaminase CRISPR base editor enables concurrent adenine and cytosine editing
2
1
9
For the few of you who follow me (love you guys! ๐ฅฐ), you may have noticed the not-so-subtle Twitter handle change.๐
I hope you will join me on this new journey to solve the biggest problems in biotech! Let's do this together. ๐ช๐พ.
0
0
11