Engineering of CRISPR-Cas PAM recognition using deep learning of vast evolutionary data

Nayfach, S.; Bhatnagar, A.; Novichkov, A.; Estevam, G. O.; Kim, N.; Hill, E.; Ruffolo, J. A.; Silverstein, R. A.; Gallagher, J.; Kleinstiver, B. P.; Meeske, A. J.; Cameron, P.; Madani, A.

doi:10.1101/2025.01.06.631536

bioRxiv · 10.1101/2025.01.06.631536

Engineering of CRISPR-Cas PAM recognition using deep learning of vast evolutionary data

Nayfach, S.·Bhatnagar, A.·Novichkov, A.·Estevam, G. O.·Kim, N.·Hill, E.·Ruffolo, J. A.·Silverstein, R. A.·Gallagher, J.·Kleinstiver, B. P.·Meeske, A. J.·Cameron, P.·Madani, A.

View original source ↗Export metadata ↓

Abstract

CRISPR-Cas enzymes must recognize a protospacer-adjacent motif (PAM) to edit a genomic site, significantly limiting the range of targetable sequences in a genome. Machine learning-based protein engineering provides a powerful solution to efficiently generate Cas protein variants tailored to recognize specific PAMs. Here, we present Protein2PAM, an evolution-informed deep learning model trained on a dataset of over 45,000 CRISPR-Cas PAMs. Protein2PAM rapidly and accurately predicts PAM specificity directly from Cas proteins across Type I, II, and V CRISPR-Cas systems. Using in silico deep mutational scanning, we demonstrate that the model can identify residues critical for PAM recognition in Cas9 without utilizing structural information. As a proof of concept for protein engineering, we employ Protein2PAM to computationally evolve Nme1Cas9, generating variants with broadened PAM recognition and up to a 50-fold increase in PAM cleavage rates compared to the wild-type under in vitro conditions. This work represents the first successful application of machine learning to achieve customization of Cas enzymes for alternate PAM recognition, paving the way for personalized genome editing.

Explore related subjects

synthetic biology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Nayfach, S., Bhatnagar, A., Novichkov, A., Estevam, G. O., Kim, N., Hill, E., Ruffolo, J. A., Silverstein, R. A., Gallagher, J., Kleinstiver, B. P., Meeske, A. J., Cameron, P., Madani, A.. 2025-01-06. Engineering of CRISPR-Cas PAM recognition using deep learning of vast evolutionary data. https://doi.org/10.1101/2025.01.06.631536

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Engineering phenotypic heterogeneity for functional organization in microbial populations

Engineering microbial populations to perform complex functions requires programming not only cellular behaviour but also the functional organization through which biological activities are distributed across populations. While phenotypic heterogeneity is often regarded as variability to suppress, it can also serve as a foundation for organizing specialized functions within genetically homogeneous microbial systems. Here, we introduce PROMETEO, a modular genetic-circuit framework that programs population composition through architecture-encoded regulatory and translational asymmetries. Using a library of 27 asymmetric bistable circuits, we demonstrate that circuit architecture reproducibly specifies phenotypic distributions spanning a broad range of population compositions without continuous external induction. Programmed population structures remained stable over serial propagation and were qualitatively conserved across Escherichia coli and Pseudomonas putida. Stochastic and deterministic modelling accurately predicted architecture-dependent population compositions and hysteresis regimes, providing a quantitative framework for rational design. We further show that programmable population composition supports multiple modes of functional organization, including stable parallel specialization, inducible temporal redistribution of cellular states, and spatial ecological compartmentalization through biofilm-associated and planktonic subpopulations. As a demonstration of these capabilities, architecture-programmed organization enabled distributed Congo Red biotransformation through coordinated reductive and oxidative activities, achieving up to 98% dye removal in spatially compartmentalized populations. Together, these results establish population composition as a programmable property of genetic circuit design and provide a general strategy for engineering distributed functions within genetically homogeneous microbial populations.

synthetic biology↗

bioRxiv2026

Carboxysome-Inspired Protein Coacervates for Light-Driven CO2 Reduction and H2 Evolution

Efficient catalysis often requires high local concentrations of reactants and catalysts, which cells achieve through compartmentalization within organelles, such as carboxysomes, that increase the efficiency of bacterial carbon fixation. Here, we engineered a photocatalytic reaction compartment that concentrated an artificial metalloenzyme, carbon dioxide, and a photosensitizer by liquid-liquid phase separation triggered by a cationic polypeptide, deca(L-arginine) (R10). At low R10 concentrations, CoPPIX binding increases the alpha-helical structure of the otherwise disordered protein, supercharged cytochrome b5622(-22). At higher concentrations, electrostatic complexation produces spherical droplets that enrich the protein and cobalt cofactor and recruit the photosensitizer [Ru(bpy)3]2+. Under illumination, coacervation increased hydrogen evolution 1.9-fold and CO formation from carbon dioxide; 1.3-fold relative to the corresponding solution-phase protein system. Co-encapsulation of carbonic anhydrase changed the product distribution specifically in the condensed phase: CO production increased 3.2-fold, hydrogen evolution decreased from 1.51 to 0.70 mol, and CO selectivity among the detected two-electron products rose from 33% to 77%. These results demonstrate that bioinspired coacervates can stabilize reactive intermediates, enrich local substrate concentrations, and integrate multiple catalytic functions, providing a generalizable framework for programmable, light-driven synthetic organelles.

synthetic biology↗

bioRxiv2026

Biocatalytic Production of Galantamine in Yeast through Cytochrome P450 Optimization

Galantamine is a pharmaceutically relevant Amaryllidaceae alkaloid used for the treatment of Alzheimer's disease. Its structural complexity and low abundance in plants motivate the development of alternative manufacturing routes. Here, we established the first engineered yeast platform for the biocatalytic production of galantamine from 4OMe-norbelladine. Heterologous expression of the downstream pathway enzymes NtCYP96T6, NtNMT1, and NtAKR1 from Narcissus cv. Tete-a-Tete in Saccharomyces cerevisiae was complemented by optimization of cultivation temperature, carbon source, and medium pH enabling the first demonstration of galantamine production in yeast. Systematic screening of cytochrome P450 reductase and cytochrome b partners to boost NtCYP96T6 activity led to the identification of a novel reductase mined from the Narcissus pseudonarcissus transcriptome, NpCPR, which supported the highest pathway flux and yielded 7.0 {+/-} 0.6 mg/L galantamine, corresponding to a 7.6 % molar yield from 250 M 4OMe-norbelladine and an approximately 173-fold improvement over the parental strain. We further exploited this yeast cell factory for the precursor-directed biosynthesis of 7F-galantamine, highlighting the potential of pathway enzyme promiscuity to access new-to-nature GAL analogues that may be challenging to produce through conventional chemical synthesis. Together, this work establishes a foundation for microbial galantamine production and biosynthetic diversification of its pharmaceutically relevant scaffold.

synthetic biology↗