Leo Chen @LeoTZ03 Twitter profile

Last Seen Profiles

@05eZ8iWefHz2ccZ

@hanamt_331

@TlhsVball

@sareenamamat

@taoruman_ch2

@DanteFarta

@theASTARart

@react_native_eu

@meeranalrefai03

@bmb__0

@sharlot_boronze

@_E_FARAZI_

@oriko_25

@sarahhanem

@WillHess65

@renatocolaco

@ClydeHo59873967

@Retweelgend

@pengen_stw

@ClcpierceCarol

@ImoveCar

@NameSeekers

@Helmut_Lehmeyer

@LeHautBitCodeur

@iwazaki_13_fans

@Hle_____

@mrs_sara99

@alara_skr

@paul_lehmans

@mcnihdogg

@storytel_fi

@mike_merko

@temul

@penaldowski

@niplcs

@millennialprof

Leo Chen

@LeoTZ03

1 month

CRISPR-GPT: An LLM Agent for Automated Design of Gene-Editing Experiments Use Agent "to facilitate the process of selecting CRISPR systems, designing guide RNAs, recommending cellular delivery methods, drafting protocols, and designing validation experiments to confirm editing…

2

76

208

Leo Chen

@LeoTZ03

20 days

ProteinCLIP: enhancing protein language models with natural language - CLIP-like Model aligns embeddings from protein language model and language models describing protein functions - Excels in PPI, Homology, and Mutation identification

3

23

141

Leo Chen

@LeoTZ03

2 months

RNA language models predict mutations that improve RNA function A great work from Jennifer Dounda and Jamie Cate's Lab! - Use Genome Taxonomy Database (GTDB) to build the GARNET (Gtdb Acquired RNa with Environmental Temperatures) database - Train a generative GNN model using a…

3

27

137

Leo Chen

@LeoTZ03

2 months

Predictomes: A classifier-curated database of AlphaFold-modeled protein-protein interactions Interesting PPI virtual screening work based on AlphaFold-Multimer. - SPOC (Structure Prediction and Omics informed Classifier) is a random forest-based classifier that accurately…

4

43

139

Leo Chen

@LeoTZ03

1 month

Learning to design protein-protein interactions with enhanced generalization ICLR24' - PPIRef, the largest and non-redundant dataset of 3D protein–protein interactions, - PPIformer, a new SE(3)- equivariant model generalizing across diverse protein-binder variants. - Finetune…

0

24

127

Leo Chen

@LeoTZ03

14 days

Accurate Conformation Sampling via Protein Structural Diffusion - Diffold, a diffusion-based model for robust sampling of diverse protein conformations using amino acid sequences - Transforms AlphaFold2 into a diffusion model, and applies hierarchical reweighting based on…

0

26

121

Leo Chen

@LeoTZ03

24 days

PocketGen: Generating Full-Atom Ligand-Binding Protein Pockets - Co-designs the residue sequence and full-atom structure of protein pockets for binding - Uses a bilevel graph transformer to model multi-granularity (atom and residue/ligand level) and multi-aspect (intra-protein…

1

35

116

Leo Chen

@LeoTZ03

22 days

Multi-Scale Protein Language Model for Unified Molecular Modeling | ICML 2024 - ESM-AA, a multi-scale protein language model that enables unified modeling at both the residue and atom scales - Pre-training on multi-scale code-switch protein sequences that randomly unzip residues…

1

22

104

Leo Chen

@LeoTZ03

2 months

A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions A new 5′ untranslated region (UTR) Language Model! - Pretrain the model with mask prediction, 5' UTR secondary strcuture prediction, and minimum free energy prediction - Finetune [CLS]…

1

19

101

Leo Chen

@LeoTZ03

2 months

InstructPLM: Aligning Protein Language Models to Follow Protein Structure Instructions - Use an adapter to connect the Structure Encoder (ProteinMPNN and others) with pLM (ProGen2) - Outperforms ProteinMPNN in terms of Perplexity and Recovery Rate - Validate the model by…

1

18

96

Leo Chen

@LeoTZ03

2 months

Designing molecular RNA switches with Restricted Boltzmann machines - Use Restricted Boltzmann machines to design artificial SAM-I riboswitches, focusing on their aptamer domain - The designed sequences were validated through chemical probing, with approximately 30% demonstrating…

0

21

90

Leo Chen

@LeoTZ03

30 days

Accurate structure prediction of biomolecular interactions with AlphaFold 3 | Nature They use Diffusion with Transformer😲. Replaced invariant point attention with a relatively standard non-equivariant point-cloud diffusion model over all atoms

1

18

89

Leo Chen

@LeoTZ03

2 months

Preference optimization of protein language models as a multi-objective binder design paradigm Use DPO (Direct Pereference Optimization) for Peptide Binder Design with ProGPT2

1

17

87

Leo Chen

@LeoTZ03

28 days

FAPM: Functional Annotation of Proteins using Multi-Modal Models Beyond Structural Modeling - BLIP-2 for Protein Annotation - Combine ESM2 and Mistral-7B - Outperforms DeepGo series

1

30

87

Leo Chen

@LeoTZ03

13 days

Mamba for Protein Design

Damiano Sgarbossa

@damiano_sga

13 days

🚀 Excited to introduce ProtMamba! Our novel protein language model designed to facilitate protein design.🔬📄 🔧 1/n

9

53

238

0

24

77

Leo Chen

@LeoTZ03

1 month

De novo generation of multi-target compounds using deep generative chemistry @NatureComms POLYGON, a VAE based model with reinforcement learning for programmatic generation of new polypharmacology compounds that inhibit multiple protein targets

0

12

75

Leo Chen

@LeoTZ03

15 days

Aligning protein generative models with experimental fitness via Direct Preference Optimization @talaldotpdb - ProteinDPO, aligns ESM-IF1 with experimental stability fitness using Direct Preference Optimization - Capable of improved binding affinity prediction and stabilized…

0

9

75

Leo Chen

@LeoTZ03

1 month

Evolution-Inspired Loss Functions for Protein Representation Learning Evolutionary Ranking (EvoRank) incorporates evolutionary dynamics from MSA-based Soft Labels to learn more diverse protein representations

1

19

82

Leo Chen

@LeoTZ03

2 months

Latent-based Directed Evolution accelerated by Gradient Ascent for Protein Sequence Design Combining Directed Evolution with Protein Language Models by Gradient Ascent One small question: It seems that Figure 2 directly uses an image from another source without proper citation.…

1

12

73

Leo Chen

@LeoTZ03

1 month

PROflow: An iterative refinement model for PROTAC-induced structure prediction - PROFLOW, a flow-matching based approach for PROTAC-induced structure prediction - Created a new pseudo-ternary dataset from binary protein-protein complexes and PROTAC linker graphs…

0

11

71

Leo Chen

@LeoTZ03

2 months

DiscDiff: Latent Diffusion Model for DNA Sequence Generation VAE-based Latent Diffusion for DNA

1

10

73

Leo Chen

@LeoTZ03

1 month

Large scale paired antibody language models IgBert and IgT5, trained on two billion unpaired sequences and two million paired sequences of light and heavy chains

1

17

71

Leo Chen

@LeoTZ03

13 days

Accurate and robust protein sequence design with CarbonDesign @NatMachIntell - Inverseformer to learn from backbone structures, a Markov random field for sequence decoding, and network recycling and multitask learning for side chain design

0

15

68

Leo Chen

@LeoTZ03

26 days

Predicting equilibrium distributions for molecular systems with deep learning @NatMachIntell "we introduce Distributional Graphormer (DiG), a new deep learning framework for predicting protein structures according to their equilibrium distribution."

1

18

68

Leo Chen

@LeoTZ03

10 days

PILOT: Equivariant diffusion for pocket conditioned de novo ligand generation with multi-objective guidance via importance sampling - PILOT, Pocket-induced Ligand Optimization Tool - Unconditional pre-training, pocket-conditioned fine-tuning, and property-guided inference…

1

17

65

Leo Chen

@LeoTZ03

11 days

A generative foundation model for antibody sequence understanding - FAbCon, a causal language model for understanding and designing antibody sequences (144M, 297M, and 2.4B) Preprint: HuggingFace Model:

0

15

64

Leo Chen

@LeoTZ03

21 days

Cramming Protein Language Model Training in 24 GPU Hours - Cramming challenge, train a transformer-based pLM from scratch in ≤24 hours on 1 GPU, using UniRef50 dataset, and evaluate on FLIP and PPI downstream tasks - Remove query/key/value biases and intermediate linear layer…

0

14

59

Leo Chen

@LeoTZ03

2 months

HelixFold-Multimer: Elevating Protein Complex Structure Prediction to New Heights HelixFold extends to Multimer case. They showed better performance against AlphaFold and RoseTTAFold

1

11

58

Leo Chen

@LeoTZ03

22 days

Protein Embeddings Predict Binding Residues in Disordered Regions - IDBindT5, ProtT5 for IDR binding sites prediction (I love the style of Figure 1😂)

2

8

57

Leo Chen

@LeoTZ03

26 days

The Continuous Language of Protein Structure - Generative Invariant Angle Transformer (GIAT), an autoregressive language model that generates protein backbone structures by sampling dihedral angles one residue at a time - Neural von Mises Mixture (NVMM) layer parameterizes the…

0

8

57

Leo Chen

@LeoTZ03

1 month

ICLR '24 GEM (Generative and Experimental Perspectives for Biomolecular Design) Workshop acceptances are now open to the public. You can find many interesting works here.

0

16

54

Leo Chen

@LeoTZ03

2 months

This is really exciting!! - The most extensive dataset of CRISPR operons curated to-date - Finetune ProGen2 to generate diverse CRISPR-Cas proteins I still remember the first protein design project assigned to me (back to 2022) was to design Cas proteins with protein language…

Jeff Ruffolo

@jeffruffolo

2 months

One of the first projects we took on at @ProfluentBio was designing novel gene editing proteins with language models. This grew into an initiative called OpenCRISPR, and today I’m excited to share that work (including the sequences we designed)!

2

54

242

1

13

52

Leo Chen

@LeoTZ03

14 days

MethylBERT: A Transformer-based model for read-level DNA methylation pattern identification and tumour deconvolution Language Model on Epigenomics? - Uses BERT for encoding methylation patterns, classifies reads into tumor or normal cells, and estimates tumor purity…

0

7

50

Leo Chen

@LeoTZ03

15 days

ProtT3: Protein-to-Text Generation for Text-based Protein Understanding - Use Q-Fomer to connect ESM2 and Galactica for Protein Text QA, Annotation and Retrival - Two stage training: Retreival and then Generation - Collect and set up new benchmark for Protein Text Evaluation…

0

9

49

Leo Chen

@LeoTZ03

1 month

Topology-Driven Negative Sampling Enhances Generalizability in Protein-Protein Interaction Prediction - Topological PPNI (Protein-Protein Non-Interaction) to sample high-quality hard negatives using PPI network topology - Contrastive-L3 (CL3) hypothesis to capture…

1

9

48

Leo Chen

@LeoTZ03

11 days

Genesis: A Modular Protein Language Modelling Approach to Immunogenicity Prediction - Genesis for predicting the immunogenicity of neoantigens - pMHC Encoding Module, TCR Encoding Module, and Context-specific Immunogenicity Prediction Heads Preprint:

0

4

47

Leo Chen

@LeoTZ03

25 days

Multi-purpose RNA language modelling with motif-aware pretraining and type-guided fine-tuning @NatMachIntell - Motif-aware pretraining, use base, subsequence and motif-level masking - Type-guided fine-tuning that predicts RNA types and guide feature embedding for downstream…

1

9

43

Leo Chen

@LeoTZ03

2 months

RiboDiffusion: Tertiary Structure-based RNA Inverse Folding with Generative Diffusion Models

1

9

41

Leo Chen

@LeoTZ03

1 month

GENERATIVE ACTIVE LEARNING FOR THE SEARCH OF SMALL-MOLECULE PROTEIN BINDERS - LAMBDAZERO, a generative active learning approach to efficiently search for synthesizable small molecules with desired properties - Applied with molecular docking to design novel small molecules that…

0

7

39

Leo Chen

@LeoTZ03

2 months

gRNAde: Geometric Deep Learning for 3D RNA inverse design A Geometric RNA Design Pipeline that Converts Conformational Ensembles into RNA Sequences by @chaitjo . It can be regarded as an equivalent tool to ProteinMPNN for protein design.

1

11

37

Leo Chen

@LeoTZ03

9 days

A protein sequence-based deep transfer learning framework for identifying human proteome-wide deubiquitinase-substrate interactions @NatureComms - TransDSI (Transfer Learning for Deubiquitinase-Substrate Interactions), protein sequence information + variational graph autoencoder…

0

6

47

Leo Chen

@LeoTZ03

17 days

UniCorn: A Unified Contrastive Learning Approach for Multi-view Molecular Representation Learning Multi-view representations: - 2D graph masking - 2D-3D contrastive learning - 3D denoising

1

6

35

Leo Chen

@LeoTZ03

9 days

Functional Protein Design with Local Domain Alignment - PAAG (Protein-Annotation Alignment Generation), a framework for protein design using textual annotations - ProtBERT for proteins and SciBERT for annotation - Align embeddings at both the domain and protein levels using…

0

5

36

Leo Chen

@LeoTZ03

14 days

Are Genomic Language Models All You Need? Exploring Genomic Language Models on Protein Downstream Tasks - gLMs perform pLMs on protein tasks like melting point prediction and stability prediction - Investigate 6-mer and 3-mer tokenization schemes impact on Performance - Highlight…

0

8

32

Leo Chen

@LeoTZ03

28 days

Scaffold-Lab: Critical Evaluation and Ranking of Protein Backbone Generation Methods in A Unified Framework Evaluated Genie, FrameDiff, RFdiffusion, TDS, FrameFlow, GPDL, and Chroma

0

5

28

Leo Chen

@LeoTZ03

1 month

Integrating Large-Scale Protein Structure Prediction into Human Genetics Research Interesting ANNUAL REVIEW

1

5

27

Leo Chen

@LeoTZ03

9 days

Decoupled Sequence and Structure Generation for Realistic Antibody Design - Two-stage approach: sequence design followed by structure prediction - Composition-based objective with REINFORCE trick to address token repetition in non-autoregressive models. Use ESM-2 (+LoRA) and…

0

6

27

Leo Chen

@LeoTZ03

1 month

Large-scale chemoproteomics expedites ligand discovery and predicts ligand behavior in cells - Chemoproteomics revealed 47,658 interactions between 407 fragments and 2600 proteins, mostly without prior ligands - Ligands developed for specific proteins, showing translational…

0

3

14

Leo Chen

@LeoTZ03

9 days

AlphaFold3, a secret sauce for predicting mutational effects on protein-protein interactions - Benchmark AlphaFold3 server over SKEMPI Preprint:

2

7

52

Leo Chen

@LeoTZ03

10 days

A lot exciting protein ML papers came out today

2

0

12

Leo Chen

@LeoTZ03

9 days

AptaGPT: Advancing aptamer design with a generative pre-trained language model - GPT-2 like model, 86 million parameters, BPE. - Uses SELEX data (35 NT length) from the 3rd, 6th, and 20th rounds of enrichment, focusing on B-cell maturation antigen (BCMA) - Train on 3rd round…

0

3

12

Leo Chen

@LeoTZ03

2 months

@json_yim I will put the two I saw here - -

Antigen-Specific Antibody Design via Direct Energy-based...

Antibody design, a crucial task with significant implications across various disciplines such as therapeutics and biology, presents considerable challenges due to its intricate nature. In this...

arxiv.org

0

2

10

Leo Chen

@LeoTZ03

9 days

Learning the Language of Antibody Hypervariability - AbMAP (Antibody Mutagenesis-Augmented Processing), a transfer learning framework + pLMs for better antibody modeling - in silico Mutagenesis and multi-tasking learning of structural and functional similarity Preprint:…

0

12

38

Leo Chen

@LeoTZ03

2 months

RNA language models predict mutations that improve RNA function

Structured RNA lies at the heart of many central biological processes, from gene expression to catalysis. While advances in deep learning enable the prediction of accurate protein structural models,...

www.biorxiv.org

0

1

7

Leo Chen

@LeoTZ03

3 months

[Transformer Model Generated Bacteriophage Genomes are Compositionally Distinct from Natural Sequences] A Generative Model for Viral Genome based on MegaByte Model from @AIatMeta

Jeremy Ratcliff

@jeremydratcliff

3 months

For my first solo author venture, I performed an in-depth stress test of the outputs of the megaDNA transformer model. Through applying fun math I offer some guidance for the intersection of language models and genomics.

2

9

56

1

2

6

Leo Chen

@LeoTZ03

17 days

Metabuli: sensitive and specific metagenomic classification via joint analysis of amino acid and DNA @naturemethods - Leverages both DNA and AA information by metamers, novel k-mers that jointly encode DNA and AA information - Finds exact AA matches and uses DNA information for…

0

7

Leo Chen

@LeoTZ03

2 months

A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions

Nature Machine Intelligence - The 5′ untranslated region is a critical regulatory region of mRNA, influencing gene expression regulation and translation. Chu, Yu and colleagues develop a...

www.nature.com

0

7

Leo Chen

@LeoTZ03

3 months

Very interesting work. Definitely worth further reading!

Anthony Gitter

@anthonygitter

3 months

Our manuscript "Biophysics-based protein language models for protein engineering" with @romerolab1 is now on bioRxiv. We present Mutational Effect Transfer Learning (METL), a protein language model trained on biophysical simulations, and showcase it for protein engineering. 1/

6

70

357

0

2

6

Leo Chen

@LeoTZ03

2 months

Diffusion on language model embeddings for protein sequence generation They trained a Diffusion Model on the Continuous Latent Space of ESM-2-8M, and it outperfoms other sequence-based generative models. It is very similar to what we have done for AMP-Diffusion (Diffusion on…

2

5

Leo Chen

@LeoTZ03

3 months

Unified Generative Modeling of 3D Molecules with Bayesian Flow Networks GeoBFN extends Bayesian Flow Network to 3D molecules generation. Geometric Bayesian Flow Networks (GeoBFN) is a new generative model for 3D molecule geometry that operates in the differentiable parameter…

0

5

Leo Chen

@LeoTZ03

2 months

Predictomes: A classifier-curated database of AlphaFold-modeled protein-protein interactions

Protein-protein interactions (PPIs) are ubiquitous in biology, yet a comprehensive structural characterization of the PPIs underlying biochemical processes is lacking. Although AlphaFold-Multimer...

www.biorxiv.org

1

3

6

Leo Chen

@LeoTZ03

2 months

Predictomes | HOME

A free interactive resource for viewing predicted AlphaFold multimer structures for protein pairs in biological pathways.

predictomes.org

0

2

6

Leo Chen

@LeoTZ03

25 days

Multi-purpose RNA language modelling with motif-aware pretraining and type-guided fine-tuning

Nature Machine Intelligence - Despite the existence of various pretrained language models for nucleotide sequence analysis, achieving good performance on a broad range of downstream tasks using a...

www.nature.com

0

5

Leo Chen

@LeoTZ03

3 months

Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking "We present Quiet-STaR, a generalization of STaR in which LMs learn to generate rationales at each token to explain future text, improving their predictions. We address key…

0

1

5

Leo Chen

@LeoTZ03

30 days

just took sequences from CASP 16 Target: M1209 and ran AF3. Here is the output. Thoughts 🤔? Did I do anything wrong?

0

4

Leo Chen

@LeoTZ03

2 months

Preference optimization of protein language models as a...

We present a multi-objective binder design paradigm based on instruction fine-tuning and direct preference optimization (DPO) of autoregressive protein language models (pLMs). Multiple design...

arxiv.org

0

5

Leo Chen

@LeoTZ03

3 months

So ManyFolds, So Little Time: Efficient Protein Structure Prediction With pLMs and MSAs A Folding prediction model that uses pLM and MSA

0

2

5

Leo Chen

@LeoTZ03

2 months

Rapid and Sensitive Protein Complex Alignment with Foldseek-Multimer Foldseek now supports Multimer Alignment! - Foldseek-Multimer allows fast querying of input complex(es) against millions of targets - It represents each chain-to-chain alignment as a superposition and uses the…

1

2

5

Leo Chen

@LeoTZ03

1 month

Totally agree. Disagreement Fuels Progress, Not Disrespect.

Gautam Kamath

@thegautamkamath

1 month

A few of my research principles (& recommend to others): -don't call the earnest contents of peoples' papers "garbage" -laud (don't shame) work for being forthcoming with flaws -accept that the 1st paper on a topic won't be uniformly SotA vs *the most popular topic in ML* 1/2

21

34

504

2

0

5

Leo Chen

@LeoTZ03

2 months

Really cool application! “In this study, we use a dataset of over 524,000 Trastuzumab variants to show that standard computational methods such as BLOSUM, AbLang, ESM, and ProteinMPNN can be used to design highly diverse, binder-enriched antibody libraries from just a single…

Lewis Chinery

@LewisChinery

2 months

Happy to share our recent work on antibody library design, affinity prediction, and optimisation - 'Baselining the Buzz. Trastuzumab-HER2 affinity, and beyond!' Preprint (inc. SI) - Code - Data - 🧵

6

17

93

1

2

5

Leo Chen

@LeoTZ03

2 months

Diffusion on language model embeddings for protein sequence generation [] AMP-Diffusion []

0

1

4

Leo Chen

@LeoTZ03

3 months

@instadeepai team has two interesting preprints for protein design. Contrasting Sequence with Structure: Pre-training Graph Representations with PLMs They use CLIP module to align protein sequence and strucure at both residue and chain level.

1

0

4

Leo Chen

@LeoTZ03

3 months

NeuroFold: A Multimodal Approach to Generating Novel Protein Variants in silico NeuroFold is a novel multimodal model that integrates sequence, structural, and evolutionary information to design functional enzyme variants, as demonstrated by the…

0

4

Leo Chen

@LeoTZ03

2 months

Nicheformer: a foundation model for single-cell and spatial omics Nicheformer Dropped!

0

4

Leo Chen

@LeoTZ03

3 months

ProLLaMA: A Protein Large Language Model for Multi-Task Protein Language Processing ProLLaMA introduces a novel training framework that transforms general LLMs into multi-task Protein LLMs capable of understanding both protein sequences and natural…

0

1

4

Leo Chen

@LeoTZ03

1 month

Identification of clinically relevant T cell receptors for personalized T cell therapy using combinatorial algorithms @NatureBiotech - TRTpred, an antigen-agnostic in silico predictor of tumor-reactive T cell receptors (TCRs), using classic machine learning techniques (elastic…

0

1

4

Leo Chen

@LeoTZ03

3 months

Machine learning is IID; science is OOD

David W Hogg

@davidwhogg

3 months

Simon Batzner (DeepMind), speaking @iaifi_news two days ago; “Machine learning is IID; science is OOD.” I couldn’t agree more.

3

22

149

1

4

Leo Chen

@LeoTZ03

2 months

@jure @ChanZuckerberg 1,000 is wild for a lab

0

4

Leo Chen

@LeoTZ03

3 months

Branched chemically modified poly(A) tails enhance the translation capacity of mRNA “… designed and synthesized topologically and chemically modified mRNAs with multiple synthetic poly(A) tails. Here we demonstrate that the optimized multitailed mRNA yielded ~4.7–19.5-fold…

0

2

4

Leo Chen

@LeoTZ03

3 months

Proteome-scale discovery of protein degradation and stabilization effectors

Nature - A synthetic proteome-scale strategy enables the identification of a diverse range of human proteins that can induce the degradation or stabilization of a target protein in a...

www.nature.com

0

1

4

Leo Chen

@LeoTZ03

3 months

FoldToken: Learning Protein Language via Vector Quantization and Beyond They trained VA-VAE Like model to learn discrect folding tokens and then a GPT-like model for generation and modeling.

0

1

4

Leo Chen

@LeoTZ03

2 months

Two papers came out on Nature discussing how FOXO1 can potentially enhance CAR-T therapy. FOXO1 enhances CAR T cell stemness, metabolic fitness and efficacy FOXO1-ADA (a constitutively active variant of FOXO1) overexpression leads to increased polyfunctionality of CAR T cells,…

1

3

Leo Chen

@LeoTZ03

2 months

Structural basis of Integrator-dependent RNA polymerase II termination

0

1

3

Leo Chen

@LeoTZ03

2 months

Bilingual Language Model for Protein Sequence and Structure Seq2Seq Strucutre (FoldSeek Tokens) and Sequence co-training

0

1

3

Leo Chen

@LeoTZ03

2 months

DiscDiff: Latent Diffusion Model for DNA Sequence Generation

This paper introduces a novel framework for DNA sequence generation, comprising two key components: DiscDiff, a Latent Diffusion Model (LDM) tailored for generating discrete DNA sequences, and...

arxiv.org

0

1

3

Leo Chen

@LeoTZ03

2 months

HelixFold-Multimer: Elevating Protein Complex Structure Prediction...

While monomer protein structure prediction tools boast impressive accuracy, the prediction of protein complex structures remains a daunting challenge in the field. This challenge is particularly...

arxiv.org

0

1

3

Leo Chen

@LeoTZ03

3 months

Continuous evolution of compact protein degradation tags regulated by selective molecular glues

0

3

Leo Chen

@LeoTZ03

16 days

@talaldotpdb @BrianHie @rm_rafailov Let’s goooo! 🔥

0

3

Leo Chen

@LeoTZ03

2 months

@json_yim So true lol. Already saw two of them

1

0

3

Leo Chen

@LeoTZ03

2 months

Engineering highly active and diverse nuclease enzymes by combining machine learning and ultra-high-throughput screening "...we combined ultra-high-throughput functional screening with ML and compared to parallel in-vitro directed evolution (DE) and in-silico hit recombination…

0

1

2

Leo Chen

@LeoTZ03

3 months

New Preprint from Baker’s Lab! “Here we demonstrate that a fine-tuned RFdiffusion network is capable of designing de novo antibody variable heavy chains (VHH's) that bind user-specified epitopes.”

Atomically accurate de novo design of single-domain antibodies

Despite the central role that antibodies play in modern medicine, there is currently no way to rationally design novel antibodies to bind a specific epitope on a target. Instead, antibody discovery...

www.biorxiv.org

0

1

3

Leo Chen

@LeoTZ03

12 days

@damiano_sga How did you make the mamba figure btw? It looks so cool :)

1

0

3

Leo Chen

@LeoTZ03

2 months

@amelie_iska @huggingface 🔥💯💯 let’s go! Maybe I should write some blog posts as well lol

1

0

2

Leo Chen

@LeoTZ03

2 months

@SimMat20 @anthonygitter @countablyfinite @LucyColwell37 It’s super cool, but I don’t think it’s his work tho?

0

Leo Chen

@LeoTZ03

3 months

PPIscreenML: Structure-based screening for protein-protein interactions using AlphaFold It uses AF2 confidence measures and energetic terms from the Rosetta scoring function to classify interacting proteins and compelling decoy pairs.

0

1

Leo Chen

@LeoTZ03

2 months

PLMSearch: Protein language model powers accurate and fast sequence search for remote homology "PLMSearch (Protein Language Model), a homologous protein search method with only sequences as input. PLMSearch uses deep representations from a pre-trained protein language model and…

0

1

2

Leo Chen

@LeoTZ03

2 months

FOXO1 is a master regulator of memory programming in CAR T cells

Nature - The transcription factor FOXO1 has a key role in human T cell memory, and manipulating FOXO1 expression could provide a way to enhance CAR T cell therapies by increasing CAR T cell...

www.nature.com

0

2

Leo Chen

@LeoTZ03

3 months

Programmable protein expression using a genetically encoded m6A sensor The study introduces GEMS, a novel genetically encoded sensor for detecting m6A RNA methylation, by producing fluorescence in its presence. This technique leverages a fusion protein…

Programmable protein expression using a genetically encoded m6A sensor

Nature Biotechnology - m6A modifications are measured in living cells with a fluorescent reporter.

www.nature.com

0

2

Leo Chen

@LeoTZ03

2 months

Model Stock: All we need is just a few fine-tuned models This paper introduces an efficient fine-tuning method for large pre-trained models, offering strong in-distribution (ID) and out-of distributions (OOD) performance

1

2

Leo Chen

@LeoTZ03

3 months

IMPROVING ALPHAFOLD2 PERFORMANCE WITH A GLOBAL METAGENOMIC & BIOLOGICAL DATA SUPPLY CHAIN The study introduces BaseFold, an advancement of AlphaFold2, by incorporating a comprehensive global metagenomic and biological data supply chain. This innovative…

IMPROVING ALPHAFOLD2 PERFORMANCE WITH A GLOBAL METAGENOMIC & BIOLOGICAL DATA SUPPLY CHAIN

Scaling laws suggest that more than a trillion species inhabit our planet but only a miniscule and unrepresentative fraction (less than 0.00001%) have been studied or sequenced to date. Deep learning...

www.biorxiv.org

0

2