Computational protein design
- "This Primer provides an introduction to the main approaches in computational protein design, covering both physics-based and machine-learning-based tools. It aims to be accessible to biological, physical and computer scientists alike."
www.nature.com/articles/s43...

Protein-Based Degraders: From Chemical Biology Tools to Neo-Therapeutics The nascent field of targeted protein degradation (TPD) could revolutionize biomedicine due to the ability of degrader molecules to selectively modulate disease-relevant proteins. A key limitation to ...

We describe existing platforms for protein/peptide-based ligand identification and the drug delivery systems that might be exploited for the delivery of biologic-based degraders."
Link: pubs.acs.org/doi/10.1021/...

Protein-Based Degraders: From Chemical Biology Tools to Neo-Therapeutics
- "we provide a comprehensive and critical review of studies that have used proteins and peptides to mediate the degradation and hence the functional control of otherwise challenging disease-relevant protein targets.

-- aim to approximate soft optimal denoising processes (a.k.a. policies in RL) that combine pre-trained denoising processes with value functions serving as look-ahead functions that predict from intermediate states to terminal rewards. "

- "We review these methods from a unified perspective, demonstrating that current techniques -- such as Sequential Monte Carlo (SMC)-based guidance, value-based sampling, and classifier guidance

Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Review
arxiv.org/abs/2501.09685

- Construct full-length proteins with binding motifs and refining structures using the Rosetta FastDesign protocol and grafting (with a potential round of LigandMPNN optimization)
- Engineer and validate binders for Bcl2–venetoclax, DB3–progesterone, and PDF1–actinonin through experimental testing

- Benchmark MaSIF-neosurf against RFAA on 14 ligand-induced PPI complexes with 8,907 decoys from PDBBind
- Use MaSIF-search to predict buried surfaces and identify complementary surface fingerprints from a database of protein fragments (~640,000)

Targeting protein–ligand neosurfaces with a generalizable deep learning tool - Nature A computational deep learning approach is used to design synthetic proteins that target the neosurfaces formed by protein–ligand interactions, with applications in the development of new therapeutic m...

Targeting protein–ligand neosurfaces with a generalizable deep learning tool | @Nature
- MaSIF-neosurf can design binders for protein-ligand complexes, targeting neosurfaces (i.e., ligand-induced structural changes on the protein surface)
Link: www.nature.com/articles/s41...

- Train sequence based models to predict the activity of regulatory elements (MPRALegNet, MPRAnn, EnformerMPRA, and SeiMPRA)
- Use MPRALegNet predicts TFBS combinations, fine-mapping and variant effects

Massively parallel characterization of transcriptional regulatory elements
- Develope an optimized lentiMPRA (lentiviral massively parallel reporter assay) method to test regulatory activity of >680,000 sequences across three cell types (HepG2, K562, WTC11)
Link: www.nature.com/articles/s41...

Integrating genetic algorithms and language models for enhanced enzyme design
academic.oup.com/bib/article/...

DNALONGBENCH: A Benchmark Suite for Long-Range DNA Prediction Tasks
www.biorxiv.org/content/10.1...
Engineering of CRISPR-Cas PAM recognition using deep learning of vast evolutionary data
www.biorxiv.org/content/10.1...

- "This review systematically summarizes recent advances in chromatin interaction matrix prediction models...This article details various models, focusing on how one-dimensional (1D) information transforms into the 3D structure chromatin interactions"

A review of deep learning models for the prediction of chromatin interactions with DNA and epigenomic profiles | @BriefingBioinfo
Link: academic.oup.com/bib/article/...

EnzymeCAGE: A Geometric Foundation Model for Enzyme Retrieval with Evolutionary Insights
www.biorxiv.org/content/10.1...
Semantic mining of functional de novo genes from a genomic language model
www.biorxiv.org/content/10.1...

Bridging Sequence-Structure Alignment in RNA Foundation Models
arxiv.org/abs/2407.11242
Mapping targetable sites on the human surfaceome for the design of novel binders
www.biorxiv.org/content/10.1...

NeuralPLexer3: Physio-Realistic Biomolecular Complex Structure Prediction with Flow Models
arxiv.org/abs/2412.10743
FlowDock: Geometric Flow Matching for Generative Protein-Ligand Docking and Affinity Prediction
arxiv.org/abs/2412.10966

Leveraging ancestral sequence reconstruction for protein representation learning
www.nature.com/articles/s42...
Guiding Generative Protein Language Models with Reinforcement Learning
arxiv.org/abs/2412.12979

Harnessing the biology of regulatory T cells to treat disease - Nature Reviews Drug Discovery Regulatory T cells keep the immune system in check to maintain homeostasis and restrain inflammation. This Review discusses strategies to harness these cells therapeutically for autoimmunity, transpla...

Harnessing the biology of regulatory T cells to treat disease
- "This Review will discuss recent advances in our understanding of human Treg cell biology, with a focus on mechanisms of action and strategies to assess outcomes of Treg cell-targeted therapies."
www.nature.com/articles/s41...

IgDesign: In vitro validated antibody design against multiple therapeutic antigens using inverse folding
www.biorxiv.org/content/10.1...

Annotation-guided Protein Design with Multi-Level Domain Alignment
arxiv.org/abs/2404.16866
BEACON: Benchmark for Comprehensive RNA Tasks and Language Models
arxiv.org/abs/2406.10391

mRNA m6A detection - Nature Reviews Methods Primers N6-methyladenosine (m6A) is an mRNA modification influencing gene expression. Advanced methodologies for mapping m6A enhance understanding of its dynamic roles and interactions. In this Primer, Moshit...

mRNA m6A detection | @MethodsPrimers
- "This Primer outlines the available tools for detecting and mapping m6A, discusses the strengths and limitations of each method and offers guidance on selecting the most suitable approach."
www.nature.com/articles/s43...

- Use gradient-based approximation to modify protein sequences to increase/decrease specific concept values (e.g., which amino acids for increasing aromaticity).

- Train model with MLM Loss, Concept Loss (mean square error on concept embedding), and Orthogonality Loss (cosine similarity between known/unknown embeddings).

- Add Concept Bottleneck Module (using <cls> token) and Orthogonality Network to standard BERT-like architecture.

Concept Bottleneck Language Models For protein design
- Introduce CB-pLM (Concept Bottleneck Protein Language Models) from 24M to 3B, trained on UniRef50 and SwissProt over 718 concepts (including Cluster name, Biological process, and Biopython-derived features, etc.)
arxiv.org/abs/2411.06090

Benchmarking recent computational tools for DNA-binding protein identification Abstract. Identification of DNA-binding proteins (DBPs) is a crucial task in genome annotation, as it aids in understanding gene regulation, DNA replicatio

Benchmarking recent computational tools for DNA-binding protein identification
- "we conduct an unbiased benchmarking of 11 state-of-the-art computational tools as well as traditional tools such as ScanProsite, BLAST, and HMMER for identifying DBPs."
Link: academic.oup.com/bib/article/...

Title correction:

A general temperature-guided language model to design proteins of enhanced stability and activity

- Mouse level: Human-homologous protein data sourced from OGEE database
- Cell line level: Protein essentiality data from Project Score database, providing insights across 323 different human cell lines