Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Automatic detection of alignment errors in cryo-electron tomography

  • de Isidro-Gómez, F P
  • Vilas, J. L.
  • Carazo, José M.
  • Sorzano, Carlos Óscar S.
Supplementary material related to this article can be found online at https://doi.org/10.1016/j.jsb.2024.108153, Cryo-electron tomography is an imaging technique that allows the study of the three-dimensional structure of a wide range of biological samples, from entire cellular environments to purified specimens. This technique collects a series of images from different views of the specimen by tilting the sample stage in the microscope. Subsequently, this information is combined into a three-dimensional reconstruction. To obtain reliable representations of the specimen of study, it is mandatory to define the acquisition geometry accurately. This is achieved by aligning all tilt images to a standard reference scheme. Errors in this step introduce artifacts into the final reconstructed tomograms, leading to loss of resolution and making them unsuitable for detailed sample analysis. This publication presents algorithms for automatically assessing the alignment quality of the tilt series and their classification based on the residual errors provided by the alignment algorithms. If no alignment information is available, a set of algorithms for calculating the residual vectors focused on fiducial markers is also presented. This software is accessible as part of the Xmipp software package and the Scipion framework., The authors acknowledge the financial support from the Ministry of Science, Innovation, and Universities, Spain (BDNS n. 716450) to the Instruct Image Processing Center (I2PC) as part of the Spanish participation in Instruct-ERIC, the European Strategic Infrastructure Project (ESFRI) in the area of Structural Biology. This work was also supported by Grant PID2022-136594NB-I00 funded by MICIU/AEI/10.13039/501100011033 and ‘‘ERDF A way of making Europe’’, by the European Union and Comunidad Autónoma de Madrid, Spain through Grant S2022/BMD-7232. Additional support came from the European Union (EU) and Horizon 2020 through the HighResCells grant (ERC- 2018 - SyG, Proposal: 810057) and from the European Union (EU) and Horizon Europe through the Fragment Screen grant (Proposal: 101094131)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Real-space heterogeneous reconstruction, refinement, and disentanglement of CryoEM conformational states with HetSIREN

  • Herreros, David
  • Mata, Carlos P.
  • Noddings, Chari
  • Irene, Deli
  • Krieger, James
  • Agard, David A
  • Tsai, Ming-Daw
  • Sorzano, Carlos Óscar S.
  • Carazo, José M.
Single-particle analysis by Cryo-electron microscopy (CryoEM) provides direct access to the conformations of macromolecules. Traditional methods assume discrete conformations, while newer algorithms estimate conformational landscapes representing the different structural states a biomolecule explores. This work presents HetSIREN, a deep learning-based method that can fully reconstruct or refine a CryoEM volume in real space based on the structural information summarized in a conformational latent space. HetSIREN is defined as an accurate space-based method that allows spatially focused analysis and the introduction of sinusoidal hypernetworks with proven high analytics capacities. Continuing with innovations, HetSIREN can also refine the images' pose while conditioning the network with additional constraints to yield cleaner high-quality volumes, as well as addressing one of the most confusing issues in heterogeneity analysis, as it is the fact that structural heterogeneity estimations are entangled with pose estimation (and to a lesser extent with CTF estimation) thanks to its decoupling architecture., The authors acknowledge the financial support from the Ministry of Science, Innovation and Universities (BDNS n. 716450) to the Instruct Image Processing Center (I2PC) as part of the Spanish participation in Instruct-ERIC, the European Strategic Infrastructure Project (ESFRI) in the area of Structural Biology, Grant [PID2022-136594NB-I00] (COSS and JMC) funded by MICIU /AEI/10.13039/501100011033/ and “ERDF A way of making Europe”, by the “European Union”, Spanish State Research Agency AEI/10.13039/501100011033, through the “Severo Ochoa” Programme for Centres of Excellence in R\&D [CEX2023-001386-S] (COSS and JMC), Comunidad Autónoma de Madrid” through Grant: S2022/BMD-7232 (COSS and JMC), European Union (EU) and Horizon 2020 through grant: HighResCells (ERC - 2018 - SyG, Proposal: 810057) (COSS and JMC) European Union (EU), and Horizon Europe through grant: Fragment Screen Proposal: 101094131 (COSS and JMC). We thank Medigen Vaccine Biologics Corp. (MVC, No. 68, Shengyi 3rd Rd., Zhubei City, Hsinchu County 302, Taiwan) for providing the SARS-CoV-2 spike protein used in this work. The SARS-CoV-2 spike protein cryo-EM work was supported by Academia Sinica (AS-KPQ-109-TPP2) (MDT) and Taiwan Cryo-EM Consortium (grant no. NSTC 113-2740-B-006-004) (MDT). Cryo-EM experiments were performed at the Academia Sinica Cryo-EM Facility supported by Academia Sinica (AS-CFII-108-110) (MDT), and initial image processing steps were performed at the Academia Sinica Grid-computing Center. We also thank Dr. Mohamad Harastani for helping generate the adenylate kinase dataset., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

How many (distinguishable) classes can we identify in single-particle analysis?

  • Lauzirika, O
  • Pernica, M
  • Herreros, David
  • Ramírez-Aportela, Erney
  • Krieger, James
  • Gragera, Marcos
  • Iceta, Mikel
  • Conesa Mingo, Pablo
  • Fonseca, Y.
  • Jiménez, Jorge
  • Filipovic, J
  • Carazo, José M.
  • Sorzano, Carlos Óscar S.
Heterogeneity in cryoEM is essential for capturing the structural variability of macromolecules, reflecting their functional states and biological significance. However, estimating heterogeneity remains challenging due to particle misclassification and algorithmic biases, which can lead to reconstructions that blend distinct conformations or fail to resolve subtle differences. Furthermore, the low signal-to-noise ratio inherent in cryo-EM data makes it nearly impossible to detect minute structural changes, as noise often obscures subtle variations in macromolecular projections. In this paper, we investigate the use of p-values associated with the null hypothesis that the observed classification differs from a random partition of the input data set, thereby providing a statistical framework for determining the number of distinguishable classes present in a given data set., The authors acknowledge financial support from the Ministry of Science, Innovation and Universities (BDNS No. 716450) to Instruct–Spain, as part of Spain's participation in Instruct–ERIC, the European Strategic Infrastructure Project (ESFRI) in Structural Biology. This work was also supported by Project PID2022-136594NB-I00, funded by MICIU/AEI/10.13039/501100011033 and the European Regional Development Fund (ERDF) under the slogan `A way of making Europe', as well as by the Spanish State Research Agency (AEI/10.13039/501100011033) through the `Severo Ochoa' Centres of Excellence in R&D Programme (Project CEX2023-001386-S). The Comunidad Autónoma de Madrid provided additional support through Project S2022/BMD-7232, and the European Union under the Horizon Europe Programme through the following projects: Fragment-Screen (Project 101094131), Harmony EuroHPC (Project 101163317), EOSC Beyond (Project 101131875) and EnLaCES (Project 101024130). The project that gave rise to these results received the support of a fellowship from `la Caixa' Foundation (ID 100010434). The fellowship code is LCF/BQ/DR24/1208003., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Merging conformational landscapes in a single consensus space with FlexConsensus algorithm

  • Herreros, David
  • Pérez Mata, Carlos
  • Sorzano, Carlos Óscar S.
  • Carazo, Jose María
Structural heterogeneity analysis in cryogenic electron microscopy is experiencing a breakthrough in estimating more accurate, richer and interpretable conformational landscapes derived from experimental data. The emergence of new methods designed to tackle the heterogeneity challenge reflects this new paradigm, enabling users to gain a better understanding of protein dynamics. However, the question of how intrinsically different heterogeneity algorithms compare remains unsolved, which is crucial for determining the reliability, stability and correctness of the estimated conformational landscapes. Here, to overcome the previous challenge, we introduce FlexConsenus: a multi-autoencoder neural network able to learn the commonalities and differences among several conformational landscapes, enabling them to be placed in a shared consensus space with enhanced reliability. The consensus space enables the measurement of reproducibility in heterogeneity estimations, allowing users to either focus their analysis on particles with a stable estimation of their structural variability or concentrate on specific particle subsets detected by only certain methods., The authors acknowledge the financial support from the Ministry of Science, Innovation, and Universities (BDNS no. 716450) to Instruct-ES as part of the Spanish participation in Instruct-ERIC; the European Strategic Infrastructure Project (ESFRI) in the area of Structural Biology, grant PID2022-136594NB-I00 (J.M.C. and C.O.S.S.), funded by MICIU /AEI/10.13039/501100011033/ and ‘ERDF A way of making Europe’ by the European Union; the Spanish State Research Agency AEI/10.13039/501100011033, through the Severo Ochoa Programme for Centres of Excellence in R&D (CEX2023-001386-S) (J.M.C. and C.O.S.S.); Comunidad Autónoma de Madrid through grant S2022/BMD-7232 (J.M.C. and C.O.S.S.); the European Union and Horizon 2020 through grant HighResCells (ERC - 2018 - SyG, Proposal: 810057) (J.M.C. and C.O.S.S.); and the European Union and Horizon Europe through grant Fragment Screen Proposal: 101094131 (J.M.C. and C.O.S.S.). We thank S. Subramaniam for providing us with the data processed and presented in this work and for his support and feedback., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Otros (Other)

From fragments to drugs: How AI and in silico methods accelerate fragment-to-lead optimization

  • Aledavood, Elnaz
  • Born, Jannis
  • Gil, Carmen
Fragment-based drug design (FBDD) has become a key approach for mapping chemical space by starting with small, low-affinity fragments and building them into potent leads. Although these fragments bind weakly, strategies like fragment merging and linking can convert them into high-affinity molecules. More recently, Artificial Intelligence (AI) and Machine Learning (ML) have sped up this workflow by powering structure-guided optimization and generative design of new compounds. This review presents the state of FBDD: from biophysical screening and rapid structure elucidation to fragment growing, merging, and linking that elevate affinity while preserving ligand efficiency and drug-like properties. In particular, this review focuses on the integration of AI/ML approaches within these workflows. We compare experimental and computational strategies, summarize representative case studies, and assess how AI/ML now supports hit triage, compound priorization, and efficient exploration of chemical space. By emphasizing the role of AI-driven methods within established FBDD pipelines, we aim to provide a perspective on how these approaches complement traditional strategies. Limitations, common artifacts, and validation practices are discussed to clarify what reliably works and where open challenges remain in FBDD, This research was supported by the EU project Fragment-Screen, grant agreement ID 101094131, funded by the Horizon Europe pro gram of the European Commission. CIB–CSIC is an accredited Chemistry Partner Site of EU-OPENSCREEN ERIC, which is hereby acknowledged., Peer reviewed, Modelo experimental alternativo-NAM, in silico
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

AlignPCA-2D: PCA-reduced Euclidean vector alignment for 2D classification in cryo-EM

  • Ramírez-Aportela, Erney
  • Lauzirika Zarrabeitia, Oier
  • Fonseca-Reyna, Yunior C.
  • Ceska, T.
  • Subramaniam, Sriram
  • Carazo, José M.
  • Sorzano, Carlos Óscar S.
Cryogenic electron microscopy (cryo-EM) has transformed structural biology by enabling the high-resolution reconstruction of macromolecular complexes from noisy projection images. However, the intrinsic heterogeneity and low signal-to-noise ratio of cryo-EM datasets make 2D classification a critical and computationally demanding step in the processing workflow. Here, we introduce AlignPCA-2D, a principal component analysis (PCA)-space Euclidean vector alignment method for fast, interpretable 2D classification in cryo-EM. By projecting particle images and class representations into a compressed latent PCA space, AlignPCA-2D reduces data dimensionality while preserving meaningful structural variability. The image-to-class assignment is then performed using Euclidean distance, enabling efficient and accurate classification. We benchmark AlignPCA-2D against established cryo-EM software, such as RELION and CryoSPARC, and demonstrate that it achieves competitive alignment accuracy while substantially reducing computational cost. This approach provides a lightweight alternative for large-scale 2D classification tasks, and its modular design makes it compatible with existing cryo-EM processing pipelines., The authors acknowledge financial support from the Ministry of Science, Innovation and Universities (BDNS/716450) to Instruct Spain as part of the Spanish participation in Instruct–ERIC, the European Strategic Infrastructure Project (ESFRI) in the area of Structural Biology, Grant PID2022-136594NB-I00 funded by MICIU/AEI/10.13039/501100011033/ and `ERDF – A way of making Europe', by the European Union, Comunidad Autónoma de Madrid through grant S2022/BMD-7232. OLZ acknowledges support from a PhD fellowship funded by the `la Caixa' Foundation (ID 100010434; fellowship code LCF/BQ/DR24/1208003). Additional support was provided by Gandeeva Therapeutics through contract 20220678., With funding from the Spanish government through the ‘Severo Ochoa Centre of Excellence’ accreditation (CEX2023-001386-S)., Peer reviewed