Biology subjects

Krishna, R.

Publications and source records attributed to Krishna, R..

At least 19 recordsLinked to original sources

bioRxiv2026

Programming N-linked glycan composition through de novo protein design

N-linked glycans are ubiquitous post-translational modifications that mediate diverse biological processes, including cell signaling and adhesion, molecular recognition, and host-pathogen interactions. Despite their importance, N-linked glycans have largely remained beyond the reach of biomolecular design due to a lack of methods capable of jointly modeling glycans and protein structure. Here, we develop a general all-atom computational pipeline for creating de novo glycoproteins in which the amino acid sequence encodes both the location and composition of N-linked glycans. The method uses three-dimensional protein structure to tune glycan accessibility to processing enzymes in the eukaryotic secretory pathway. Systematic increases in steric restriction progressively reduced processing, shifting glycans from heterogeneous complex-type structures toward increasingly underprocessed oligomannose-type species resembling those found on many viral glycoproteins. Despite this programmed restriction, the resulting oligomannose glycans remained accessible for biological recognition, inhibiting uropathogenic Escherichia coli adhesion to human bladder cells and binding mannose-binding lectin. These results establish steric control of N-glycan processing as a genetically encodable design principle and extend computational protein design to functional glycoproteins with programmed composition.

biochemistry↗

bioRxiv2026

Generative design of sequence specific DNA binding proteins

De novo protein design has advanced rapidly in recent years, yet the programmable recognition of specific DNA sequences remains a longstanding challenge. Here we describe a deep learning based approach for designing sequence selective DNA binding proteins. Our method combines structure generation using RFdiffusion3 with explicit screening against off-target interactions using AlphaFold3. We test this approach by generating 96 designs for each of 15 diverse DNA targets and identify specific binders for 7 targets, representing a ~100-fold improvement in success rates over previous approaches. We further characterize the binding landscape using variant competition assays and randomized library screening, revealing robust sequence discrimination across diverse targets. Together, these results represent a significant step forward in de novo sequence specific DNA binder design.

synthetic biology↗

bioRxiv2026

RNA foundation models enable generalizable endometriosis disease classification and stable gene-level interpretation

Endometriosis is a chronic inflammatory condition with significant diagnostic delays impacting one in ten reproductive age women worldwide. While machine learning (ML) models trained on transcriptomic data show promise for disease prediction, limited generalizability across independent patient cohorts has hindered clinical translation. Foundations models (FMs) pretrained on large-scale transcriptomic data offer promise to learn transferrable, biologically meaningful representations that could support cross-cohort predictions. We assembled a 12-cohort bulk RNA-seq benchmark (334 samples) and developed a computationally efficient pipeline to test whether FMs improve endometriosis classification, an approach not previously applied to this disease. Using AutoXAI4Omics with cohort-aware validation, we compared embeddings derived from five state-of-the-art RNA FMs against TPM baselines. In cross-cohort prediction, FM embeddings significantly improved performance, achieving a weighted F1-score of 0.83 vs. 0.68 for the baseline. To allow gene-level interpretation of FM embedding models, we introduce classified-aligned integrated gradients (CA-IG), an interpretability approach aligning gene-level attributions to the downstream classifier without end-to-end finetuning. CA-IG revealed a conserved set of predictive genes from FM embeddings across cohort-validation regimes, contrasting with unstable baseline explainability, suggesting that FM embeddings prioritized transferable disease-related signal over cohort-specific effects. These genes include novel candidates that converge on biologically plausible pathways for endometriosis.

bioinformatics↗

bioRxiv2025

De novo design of phospho-tyrosine peptide binders

Phosphorylation on tyrosine is a key step in many signaling pathways. Despite recent progress in de novo design of protein binders, there are no current methods for designing binders that recognize phosphorylated proteins and peptides; this is a challenging problem as phosphate groups are highly charged, and phosphorylation often occurs within unstructured regions. Here we introduce RoseTTAFold Diffusion 2 for Molecular Interfaces (RFD2-MI), a deep generative framework for the design of binders for protein, ligand, and covalently modified protein targets. We demonstrate the power and versatility of this method by designing binders for four critical phosphotyrosine sites on three clinically relevant targets: Cluster of Differentiation 3 (CD3{varepsilon}), Epidermal Growth Factor Receptor (EGFR), Insulin Receptor (INSR) and Signal Transducer and Activator of Transcription 5 (STAT5). Experimental characterization shows that the designs bind their phosphotyrosine containing targets with affinities comparable to native binding sites and have negligible binding to non-phosphorylated targets or phosphopeptides with different sequences. X-ray crystal structures of generated binders to CD3{varepsilon} and EGFR are very close to the design models, demonstrating the accuracy of the design approach. A designed binder to an EGFR intracellular region phosphorylated upon EGF activation co-localizes with the receptor following EGF stimulation in single-particle tracking (SPT) experiments, demonstrating pY specific recognition in living cells. RFD2-MI provides a generalizable all-atom diffusion framework for probing and modulating phosphorylation-dependent signaling, and more generally, for developing research tools and targeted therapeutics against post-translationally modified proteins.

bioengineering↗

bioRxiv2025

De novo Design of All-atom Biomolecular Interactions with RFdiffusion3

Deep learning has accelerated protein design, but most existing methods are restricted to generating protein backbone coordinates and often neglect interactions with other biomolecules. We present RFdiffusion3 (RFD3), a diffusion model that generates protein structures in the context of ligands, nucleic acids and other non-protein constellations of atoms. Because all polymer atoms are modeled explicitly, conditioning the model on complex sets of atom-level constraints for enzyme design and other challenges is both simpler and more effective than previous approaches. RFD3 achieves improved performance compared to prior approaches on a range of in silico benchmarks with one tenth the computational cost. Finally, we demonstrate the broad applicability of RFD3 by designing and experimentally characterizing DNA binding proteins and cysteine hydrolases. The ability to rapidly generate protein structures guided by complex sets of atom-level constraints in the context of arbitrary non-protein atoms should further expand the range of functions attainable through protein design.

biochemistry↗

bioRxiv2025

Generating functional and multistate proteins with a multimodal diffusion transformer

Generating proteins with the full diversity and complexity of functions found in nature is a grand challenge in protein design. Here, we present ProDiT, a multimodal diffusion model that unifies sequence and structure modeling paradigms to enable the design of functional proteins at scale. Trained on sequences, 3D structures, and annotations for 214M proteins across the evolutionary landscape, ProDiT generates diverse, novel proteins that preserve known active and binding site motifs and can be successfully conditioned on a wide range of molecular functions, spanning 465 Gene Ontology terms. We introduce a diffusion sampling protocol to design proteins with multiple functional states, and demonstrate this protocol by scaffolding enzymatic active sites from carbonic anhydrase and lysozyme to be allosterically deactivated by a calcium effector. Our results showcase ProDiTs unique capacity to satisfy design specifications inaccessible to existing generative models, thereby expanding the protein design toolkit.

bioinformatics↗

bioRxiv2025

Accelerating Biomolecular Modeling with AtomWorks andRF3

Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed backbone sequence design. We use AtomWorks to train RosettaFold-3 (RF3), a structure prediction network capable of predicting arbitrary biomolecular complexes with an improved treatment of chirality that narrows the performance gap between closed-source AlphaFold3 (AF3) and existing open-source implementations. We expect that AtomWorks will accelerate the next generation of open-source biomolecular machine learning models and that RF3 will be broadly useful as a structure prediction tool. To this end, we release the AtomWorks framework (https://github.com/RosettaCommons/atomworks), together with curated training data, code and model weights for RF3 (https://github.com/RosettaCommons/modelforge) under a permissive BSD license.

biochemistry↗

bioRxiv2025

Atom level enzyme active site scaffolding using RFdiffusion2

De novo enzyme design starts from ideal active site descriptions consisting of constellations of catalytic residue functional groups around reaction transition state(s), and seeks to generate protein structures that can accurately hold the site in place. Highly active enzymes have been designed starting from such descriptions using the generative AI method RFdiffusion [1-3], but there are two current methodological limitations. First, the geometry of the active site can only be specified at the residue level, so for each catalytic residue functional group placed around the reaction transition state, the possible locations of the residue backbone must be enumerated by building side chain rotamers back from the functional group. Second, the location of the catalytic residues along the sequence must be specified in advance, which considerably limits the space of solutions which can be sampled. Here we describe a new deep generative method, Rosetta Fold diffusion 2 (RFdiffusion2), that solves both problems, enabling enzymes to be designed from sequence agnostic descriptions of functional group locations without inverse rotamer generation. We first evaluate RFdiffusion2 on an in silico enzyme design benchmark of 41 diverse active sites and find that it is able to successfully build proteins scaffolding all 41 sites, compared to 16/41 with prior state-of-the-art deep learning methods. Next, we design enzymes around three diverse catalytic sites and characterize the designs experimentally; in each case we identify active catalysts in testing less than 96 sequences. RFdiffusion2 demonstrates the potential of atomic resolution generative models for the design of de novo enzymes directly from their reaction mechanisms.

biochemistry↗

bioRxiv2025

Inherent Specificity and Mutational Sensitivity as Quantitative Metrics for RBP Binding

RNA-binding proteins (RBPs) regulate every aspect of post-transcriptional gene expression, yet our ability to compare how selectively different RBPs recognize their targets remains limited. Binding affinity, expressed as a dissociation constant, provides a universal quantity for comparing binding strength, but no equivalent metric exists for binding specificity. Here we introduce two quantitative metrics to fill this gap: inherent specificity, which measures how selectively an RBP distinguishes its strongest binding motif from all other sequences, and variation sensitivity, which measures tolerance to single nucleotide changes within that motif. Analyzing high-throughput sequencing data across 100 RBPs in vitro and 27 in cells, we find strong correspondence between in vitro and cellular measurements for sequence-driven RBPs. Domain swap CLIP experiments demonstrate that specificity can be transferred between protein contexts. Mathematical modeling and cellular competition experiments reveal that low-specificity RBPs can paradoxically sharpen the target discrimination of high-specificity partners by occupying non-preferred sites, an emergent property not predictable from affinity alone. These metrics and accompanying R packages provide a practical framework for comparing RBP binding behaviors and modeling how RBPs compete for RNA targets across the transcriptome.

molecular biology↗

bioRxiv2025

STEAM: Spatial Transcriptomics Evaluation Algorithm and Metric for clustering performance

MotivationSpatial transcriptomic technologies allow researchers to explore the diversity and specificity of gene expression within their original tissue structure. Accurately identifying regions that are spatially coherent in both gene expression and physical tissue structures is an emerging topic, but challenging due to the lack of ground truth labels which renders complicating validation of clustering consistency and reproducibility. This highlights a need for a computational evaluation framework to rigorously and unbiasedly assess clustering performance. ResultsTo address this gap, we propose STEAM (Spatial Transcriptomics Evaluation Algorithm and Metric), a user-friendly computational pipeline designed to evaluate the consistency and reliability of clustering results by leveraging machine learning classification and prediction methods, with the goal of maintaining the spatial proximity and gene expression patterns within clusters. We benchmarked STEAM on various public datasets, spanning multi-cell to single-cell resolution, as well as spatial transcriptomics and proteomics. The results highlighted its robustness and generalizability through comprehensive statistical evaluation metrics, such as Kappa score, F1 score, accuracy, and adjusted rand index. Notably, STEAM supports multi-sample training, enabling cross-replicate clustering consistency assessment. Moreover, STEAM provides practical guidance by comparing clustering results across multiple approaches; here, we evaluated four different methods, including spatial-aware and spatial-ignorant approaches. In summary, we believe that STEAM provides researchers a promising tool for evaluating clustering robustness and benchmarking clustering performance for spatial omics data, offering valuable insights to drive reproducible discoveries in spatial biology. Availability and implementationSource code and the R software tool STEAM are available from https://github.com/fanzhanglab/STEAM.

bioinformatics↗

bioRxiv2025

CellPhenoX: An eXplainable Cell-specific machine learning method to predict clinical Phenotypes using single-cell multi-omics

Single-cell technologies have enhanced our knowledge of molecular and cellular heterogeneity underlying disease. As the scale of single-cell datasets expands, linking cell-level phenotypic alterations with clinical outcomes becomes increasingly challenging. To address this, we introduce CellPhenoX, an eXplainable machine learning method to identify cell-specific phenotypes that influence clinical outcomes. CellPhenoX integrates classification models, explainable AI techniques, and a statistical framework to generate interpretable, cell-specific scores that uncover cell populations associated with relevant clinical phenotypes and interaction effects. We demonstrated the performance of CellPhenoX across diverse single-cell designs, including simulations, binary disease-control comparisons, and multi-class studies. Notably, CellPhenoX identified an activated monocyte phenotype in COVID-19, with expansion correlated with disease severity after adjusting for covariates and interactive effects. It also uncovered an inflammation-associated gradient in fibroblasts from ulcerative colitis. We anticipate that CellPhenoX holds the potential to detect clinically relevant phenotypic changes in single-cell data with multiple sources of variation, paving the way for translating single-cell findings into clinical impact.

bioinformatics↗

bioRxiv2024

Modeling protein-small molecule conformational ensembles with ChemNet

Modeling the conformational heterogeneity of protein-small molecule interactions is important for understanding natural systems and evaluating designed systems, but remains an outstanding challenge. We reasoned that while residue level descriptions of biomolecules are efficient for de novo structure prediction, for probing heterogeneity of interactions with small molecules in the folded state an entirely atomic level description could have advantages in speed and generality. We developed a graph neural network called PLACER (Protein-Ligand Atomistic Conformational Ensemble Resolver) trained to recapitulate correct atomic positions from partially corrupted input structures from the Cambridge Structural Database and the Protein Data Bank; the nodes of the graph are the atoms in the system. PLACER accurately generates structures of diverse organic small molecules given knowledge of their atom composition and bonding, and given a description of the larger protein context, builds up structures of small molecules and protein side chains for protein-small molecule docking. Because PLACER is rapid and stochastic, ensembles of predictions can be readily generated to map conformational heterogeneity. In enzyme design efforts described here and elsewhere, we find that using PLACER to assess the accuracy and pre-organization of the designed active sites results in higher success rates and higher activities; we obtain a preorganized retroaldolase with a kcat/KM of 11000 M-1min-1, considerably higher than any pre-deep learning design for this reaction. We anticipate that PLACER will be widely useful for rapidly generating conformational ensembles of small molecule and small molecule-protein systems, and for designing higher activity preorganized enzymes.

biochemistry↗

bioRxiv2024

Phylogenetically distant enzymes localized in cytosol and plastids drive citral biosynthesis in lemongrass

Except for the genetic basis of citral-forming alcohol dehydrogenases (ADHs) in Litsea cubeba tree, and biochemical studies on citral-forming enzymes from select plants, knowledge regarding in-planta biosynthesis of citral and its metabolic origin remains limited. Here, we have elucidated the functions of an ADH (CfADH1) and an aldoketo-reductase (CfAKR2b) in citral biosynthesis in lemongrass (Cymbopogon flexuosus), one of the most cultivated aromatic crops for its citral-rich essential oil. Expression of both CfADH1 and CfAKR2b showed correlation with citral accumulation in different developmental stages. Recombinant CfADH1 and CfAKR2b, despite their sequence unrelatedness, exhibited similar kinetic properties and formed citral from geraniol with NADP cofactor. Virus-induced gene silencing in lemongrass, and transient expression in lemon balm (Melissa officinalis), demonstrated the in-planta involvement of CfADH1 and CfAKR2b in citral biosynthesis. While CfADH1 exhibited a dual cytosolic/plastidial localization, CfAKR2b was localized to cytosol. Moreover, feeding lemongrass seedlings with mevalonate- and methylerythritol-phosphate-pathway specific inhibitors combined with volatile profiling supported the role of both pathways in citral formation. Our results demonstrate phylogenetically distant enzymes localized in cytosol and plastids drive citral biosynthesis in lemongrass, indicating an evolutionary scenario aimed at maximizing the utilization of precursor pools from both cytosolic and plastidial pathways for high citral production.

plant biology↗

bioRxiv2023

Uncover spatially informed shared variations for single-cell spatial transcriptomics with STew

MotivationThe recent spatial transcriptomics (ST) technologies have enabled characterization of gene expression patterns and spatial information, advancing our understanding of cell lineages within diseased tissues. Several analytical approaches have been proposed for ST data, but effectively utilizing spatial information to unveil the shared variation with gene expression remains a challenge. ResultsWe introduce STew, a Spatial Transcriptomic multi-viEW representative learning method, to jointly analyze spatial information and gene expression in a scalable manner, followed by a data-driven statistical framework to measure the goodness of model fit. Through benchmarking using Human DLPFC data with true manual annotations, STew achieved superior performance in both clustering accuracy and continuity of identified spatial domains compared with other methods. STew is also robust to generate consistent results insensitive to model parameters, including sparsity constraints. We next applied STew to various ST data acquired from 10x Visium and Slide-seqV2, encompassing samples from both mouse and human brain, which revealed spatially informed cell type clusters. We further identified a pro-inflammatory fibroblast spatial niche using ST data from psoriatic skins. Hence, STew is a generalized method to identify both spatially informed clusters and disease-relevant niches in complex tissues. AvailabilitySource code and the R software tool STew are available from github.com/fanzhanglab/STew. Contactfan.3.zhang@cuanschutz.edu Supplementary informationSupplementary data are provided.

bioinformatics↗

bioRxiv2023

Generalized Biomolecular Modeling and Design with RoseTTAFold All-Atom

Although AlphaFold2 (AF2) and RoseTTAFold (RF) have transformed structural biology by enabling high-accuracy protein structure modeling, they are unable to model covalent modifications or interactions with small molecules and other non-protein molecules that can play key roles in biological function. Here, we describe RoseTTAFold All-Atom (RFAA), a deep network capable of modeling full biological assemblies containing proteins, nucleic acids, small molecules, metals, and covalent modifications given the sequences of the polymers and the atomic bonded geometry of the small molecules and covalent modifications. Following training on structures of full biological assemblies in the Protein Data Bank (PDB), RFAA has comparable protein structure prediction accuracy to AF2, excellent performance in CAMEO for flexible backbone small molecule docking, and reasonable prediction accuracy for protein covalent modifications and assemblies of proteins with multiple nucleic acid chains and small molecules which, to our knowledge, no existing method can model simultaneously. By fine-tuning on diffusive denoising tasks, we develop RFdiffusion All-Atom (RFdiffusionAA), which generates binding pockets by directly building protein structures around small molecules and other non-protein molecules. Starting from random distributions of amino acid residues surrounding target small molecules, we design and experimentally validate proteins that bind the cardiac disease therapeutic digoxigenin, the enzymatic cofactor heme, and optically active bilin molecules with potential for expanding the range of wavelengths captured by photosynthesis. We anticipate that RFAA and RFdiffusionAA will be widely useful for modeling and designing complex biomolecular systems.

biochemistry↗

bioRxiv2021

Interpreting machine learning models to investigate circadian regulation and facilitate exploration of clock function

The circadian clock is an important adaptation to life on earth. Here, we use machine learning to predict complex temporal circadian gene expression patterns in Arabidopsis. Most significantly, we classify circadian genes using DNA sequence features generated from public genomic resources, with no experimental work or prior knowledge needed. We use model explanation to rank DNA sequence features, observing transcript-specific combinations of potential circadian regulatory elements that discriminate temporal phase of expression. Model interpretation/explanation provides the backbone of our methodological advances, giving insight into biological processes and experimental design. Next, we use model interpretation to optimize sampling strategies when we predict circadian transcripts using reduced numbers of transcriptomic timepoints, saving both time and money. Finally, we predict the circadian time from a single transcriptomic timepoint, deriving novel marker transcripts that are most impactful for accurate prediction, this could facilitate the identification of altered clock function from existing datasets.

genomics↗

bioRxiv2020

Functional analysis of the fatty acid and alcohol metabolism of Pseudomonas putida using RB-TnSeq

ABSTRACTWith its ability to catabolize a wide variety of carbon sources and a growing engineering toolkit, Pseudomonas putida KT2440 is emerging as an important chassis organism for metabolic engineering. Despite advances in our understanding of this organism, many gaps remain in our knowledge of the genetic basis of its metabolic capabilities. These gaps are particularly noticeable in our understanding of both fatty acid and alcohol catabolism, where many paralogs putatively coding for similar enzymes co-exist making biochemical assignment via sequence homology difficult. To rapidly assign function to the enzymes responsible for these metabolisms, we leveraged Random Barcode Transposon Sequencing (RB-TnSeq). Global fitness analyses of transposon libraries grown on 13 fatty acids and 10 alcohols produced strong phenotypes for hundreds of genes. Fitness data from mutant pools grown on varying chain length fatty acids indicated specific enzyme substrate preferences, and enabled us to hypothesize that DUF1302/DUF1329 family proteins potentially function as esterases. From the data we also postulate catabolic routes for the two biogasoline molecules isoprenol and isopentanol, which are catabolized via leucine metabolism after initial oxidation and activation with CoA. Because fatty acids and alcohols may serve as both feedstocks or final products of metabolic engineering efforts, the fitness data presented here will help guide future genomic modifications towards higher titers, rates, and yields.IMPORTANCE To engineer novel metabolic pathways into P. putida, a comprehensive understanding of the genetic basis of its versatile metabolism is essential. Here we provide functional evidence for the putative roles of hundreds of genes involved in the fatty acid and alcohol metabolism of this bacterium. These data provide a framework facilitating precise genetic changes to prevent product degradation and channel the flux of specific pathway intermediates as desired.Competing Interest StatementJ.D.K. has financial interests in Amyris, Lygos, Demetrix, Napigen, Maple Bio, and Apertor Labs. C.B.E has a financial interest in Perlumi Chemicals.View Full Text

microbiology↗

bioRxiv2019

Recovering individual haplotypes and a contiguous genome assembly from pooled long read sequencing of the diamondback moth (Lepidoptera: Plutellidae)

BackgroundRecent advances in genomics have addressed the challenge that divergent haplotypes pose to the reconstruction of haploid genomes. However for many organisms, the sequencing of either field-caught individuals or a pool of heterogeneous individuals is still the only practical option. Here we present methodological approaches to achieve three outcomes from pooled long read sequencing: the generation of a contiguous haploid reference sequence, the sequences of heterozygous haplotypes; and reconstructed genomic sequences of individuals related to the pooled material. ResultsPacBio long read sequencing, Dovetail Hi-C scaffolding and linkage map integration yielded a haploid chromosome-level assembly for the diamondback moth (Plutella xylostella), a global pest of Brassica crops, from a pool of related individuals. The final assembly consisted of 573 scaffolds, with a total assembly size of 343.6Mbp a scaffold N50 value of 11.3Mbp (limited by chromosome size) and a maximum scaffold size of 14.4Mbp. This assembly was then integrated with an existing RAD-seq linkage map, anchoring 95% of the assembled sequence to defined chromosomal positions. ConclusionsWe describe an approach to resolve divergent haplotype sequences and describe multiple validation approaches. We also reconstruct individual genomes from pooled long-reads, by applying a recently developed k-mer binning method.

genomics↗