Biology subjects

Giri, N.

Publications and source records attributed to Giri, N..

7 recordsLinked to original sources

bioRxiv2025

A Labeled Dataset for AI-based Cryo-EM Map Enhancement

Cryo-electron microscopy (cryo-EM) has transformed structural biology by enabling near-atomic resolution imaging of macromolecular complexes. However, cryo-EM density maps suffer from intrinsic noise arising from structural sources, shot noise, and digital recording, which complicates accurate atomic structure building. While various methods for denoising cryo-EM density maps exist, there is a lack of standardized datasets for benchmarking artificial intelligence (AI) approaches. Here, we present an open-source dataset for cryo-EM density map denoising comprising 650 high-resolution (1-4 [A]) experimental maps paired with three types of generated label maps: regression maps capturing idealized density distributions, binary classification maps distinguishing structural elements from background, and atom-type classification maps. Each map is standardized to 1 [A] voxel size and validated through Fourier Shell Correlation analysis, demonstrating substantial resolution improvements in label maps compared to experimental maps. This resource bridges the gap between structural biology and artificial intelligence communities, enabling researchers to develop and benchmark innovative methods for enhancing cryo-EM density maps.

bioinformatics↗

bioRxiv2025

Atomic Protein Structure Modeling from Cryo-EM Using Multi-Modal Deep Learning and AlphaFold3

Cryo-electron microscopy (cryo-EM) has revolutionized structural biology by enabling near-atomic resolution visualization of protein structures. However, accurately modeling 3D atomic structures from cryo-EM density maps remains challenging, particularly for large multi-chain complexes and assemblies. Here, we present an automated pipeline that integrates multi-modal deep learning and advanced structure prediction techniques to improve model accuracy. Our approach leverages a deep learning model that combines sequence-based features from a Protein Language Model with cryo-EM density maps, enabling a richer feature representation across modalities. The deep learning-predicted voxels are utilized to build a Hidden Markov Model (HMM) and a tailored Viterbi algorithm is used to align sequences to generate an initial protein backbone structures. These backbone models then serve as templates for AlphaFold3, which refines the structures for improved accuracy. Our approach combines cryo-EM data with AlphaFold3 predictions, helping to refine and improve AlphaFold3s predicted structures. By integrating both methods, we can generate more accurate and reliable atomic models, particularly for large proteins with complex conformations.

bioinformatics↗

bioRxiv2024

De Novo Atomic Protein Structure Modeling for Cryo-EMDensity Maps Using 3D Transformer and Hidden MarkovModel

Accurately building three-dimensional (3D) atomic structures from 3D cryo-electron microscopy (cryo-EM) density maps is a crucial step in the cryo-EM-based determination of the structures of protein complexes. Despite improvements in the resolution of 3D cryo-EM density maps, the de novo conversion of density maps into 3D atomic structures for protein complexes that do not have accurate homologous or predicted structures to be used as templates remains a significant challenge. Here, we introduce Cryo2Struct, a fully automated ab initio cryo-EM structure modeling method that utilizes a 3D transformer to identify atoms and amino acid types in cryo-EM density maps first, and then employs a novel Hidden Markov Model (HMM) to connect predicted atoms to build backbone structures of proteins. Tested on a standard test dataset of 128 cryo-EM density maps with varying resolutions (2.1 - 5.6 [A]) and different numbers of residues (730 - 8,416), Cryo2Struct built substantially more accurate and complete protein structural models than the widely used ab initio method - Phenix in terms of multiple evaluation metrics. Moreover, on a new test dataset of 500 recently released density maps with varying resolutions (1.9 - 4.0 [A]) and different numbers of residues (234 - 8,828), it built more accurate models than on the standard dataset. And its performance is rather robust against the change of the resolution of density maps and the size of protein structures.

bioinformatics↗

bioRxiv2023

Cryo2Struct: A Large Labeled Cryo-EM Density MapDataset for AI-based Reconstruction of ProteinStructures

The advent of single-particle cryo-electron microscopy (cryo-EM) has brought forth a new era of structural biology, enabling the routine determination of large biological molecules and their complexes at atomic resolution. The high-resolution structures of biological macromolecules and their complexes significantly expedite biomedical research and drug discovery. However, automatically and accurately building atomic models from high-resolution cryo-EM density maps is still time-consuming and challenging when template-based models are unavailable. Artificial intelligence (AI) methods such as deep learning trained on limited amount of labeled cryo-EM density maps generate inaccurate atomic models. To address this issue, we created a dataset called Cryo2StructData consisting of 7,600 preprocessed cryo-EM density maps whose voxels are labelled according to their corresponding known atomic structures for training and testing AI methods to build atomic models from cryo-EM density maps. It is larger and of higher quality than any existing, publicly available dataset. We trained and tested deep learning models on Cryo2StructData to make sure it is ready for the large-scale development of AI methods for building atomic models from cryo-EM density maps.

bioinformatics↗

bioRxiv2023

Combining pairwise structural similarity and deep learning interface contact prediction to estimate protein complex model accuracy in CASP15

Estimating the accuracy of quaternary structural models of protein complexes and assemblies (EMA) is important for predicting quaternary structures and applying them to studying protein function and interaction. The pairwise similarity between structural models is proven useful for estimating the quality of protein tertiary structural models, but it has been rarely applied to predicting the quality of quaternary structural models. Moreover, the pairwise similarity approach often fails when many structural models are of low quality and similar to each other. To address the gap, we developed a hybrid method (MULTICOM_qa) combining a pairwise similarity score (PSS) and an interface contact probability score (ICPS) based on the deep learning inter-chain contact prediction for estimating protein complex model accuracy. It blindly participated in the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) in 2022 and ranked first out of 24 predictors in estimating the global accuracy of assembly models. The average per-target correlation coefficient between the model quality scores predicted by MULTICOM_qa and the true quality scores of the models of CASP15 assembly targets is 0.66. The average per-target ranking loss in using the predicted quality scores to rank the models is 0.14. It was able to select good models for most targets. Moreover, several key factors (i.e., target difficulty, model sampling difficulty, skewness of model quality, and similarity between good/bad models) for EMA are identified and analayzed. The results demonstrate that combining the multi-model method (PSS) with the complementary single-model method (ICPS) is a promising approach to EMA. The source code of MULTICOM_qa is available at https://github.com/BioinfoMachineLearning/MULTICOM_qa.

bioinformatics↗

bioRxiv2022

A Deep Learning Bioinformatics Approach to Modeling Protein-Ligand Interaction with cryo-EM Data in 2021 Ligand Model Challenge

Elucidating protein-ligand interaction is crucial for studying the function of proteins and compounds in an organism and critical for drug discovery and design. The problem of protein-ligand interaction is traditionally tackled by molecular docking and simulation, which is based on physical forces and statistical potentials and cannot effectively leverage cryo-EM data and existing protein structural information in the protein-ligand modeling process. In this work, we developed a deep learning bioinformatics pipeline (DeepProLigand) to predict protein-ligand interactions from cryo-EM density maps of proteins and ligands. DeepProLigand first uses a deep learning method to predict the structure of proteins from cryo-EM maps, which is averaged with a reference (template) structure of the proteins to produce a combined structure to add ligands. The ligands are then identified and added into the structure to generate a protein-ligand complex structure, which is further refined. The method based on the deep learning prediction and template-based modeling was blindly tested in the 2021 EMDataResource Ligand Challenge and was ranked first in fitting ligands to cryo-EM density maps.This results demonstrate that the deep learning bioinformatics approach is a promising direction to model protein-ligand interaction on cryo-EM data using prior structural information. The source code, data, and instruction to reproduce the results are available on GitHub repository : https://github.com/jianlin-cheng/DeepProLigand

bioinformatics↗

bioRxiv2022

Yeast galactokinase in closed conformation can switch between catalytic and signal transducer states.

S.cerevisiae galactokinase (ScGal1p), in closed conformation catalyzes the phosphorylation of galactose to galactose 1-phopshate using ATP as the phosphate donor as well as allosterically activates the GAL genetic switch in response to galactose and ATP as ligands. How both kinase and signaling activities of ScGal1p are associated with closed conformation of the protein is not understood. Conformational sampling of ScGal1p indicated that this protein samples closed kinase and closed non-kinase conformers. Closed non-kinase conformers are catalytically incompetent to phosphorylate galactose and act as a bonafide signal transducer. It was observed that toggling of side chain of highly conserved K266 of ScGal1p between S171and catalytic base D217 is responsible for transitioning of ScGal1p between signal transducer and kinase states. Interestingly in ScGal3p, the paralog of ScGal1p, which has only signal transduction activity and lacks kinase activity, a H bond between a non-conserved Y433, and a highly conserved Y57, gets broken during MD simulation. The corresponding H-bond present in ScGal1p between residues Y441 and Y63 respectively, remains intact throughout the simulations of ScGal1p.Therefore, we predicted that K266 and Y441 have a role in bifunctionality of ScGal1p. To test the above predictions, we monitored the signaling and kinase activity of ScGal1K266Rp and ScGal1Y441Ap variants. Signaling activity increased in both ScGal1Y441Ap and ScGal1K266Rp variants as compared to ScGal1wtp, whereas the kinase activity increased in ScGal1Y441Ap, but decreased in ScGal1K266Rp Based on the above, we propose that K266 and Y441 are crucial for conferring bifunctionality to ScGal1p. Author summaryGalactokinase of S.cerevisiae(ScGal1p), the first enzyme of Leloir pathway of galactose metabolism, phosphorylates galactose using ATP as the phosphate donor. ScGal1p also functions as a signal transducer of GAL regulon wherein galactose and ATP allosterically activate galactokinase. The active form of galactokinase, then sequesters the repressor ScGal80p, to activate the GAL switch. ScGal1p has a single site each for binding to galactose and ATP. How ScGal1p, a monomeric protein, performs the above two mutually exclusive activities using the same set of substrates/ligands, with the same site acting as the active site for enzymatic activity as well as allosteric site for signal transduction activity is unclear. Our findings are that this protein has a distinct conformational state for functioning as a signal transducer and a distinct conformational state for functioning as a kinase. A highly conserved lysine residue (K266) present only in fungal galactokinases, triggers the interconversion between catalysis and signal transduction states. This interconversion is subdued by H bond between Y441 and Y63. These studies suggest that the two activities of ScGal1p are fine tuned by evolution to regulate metabolism through transcriptional control.

genetics↗