Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures

Husain, Syed Sameed; Ong, Eng-Jon; Dmitry, Minskiy,; Bober-Irizar, Mikel; Irizar, Amaia; Bober, Miroslaw

doi:10.1038/s42003-023-04840-z

Cited by 6 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Future research is planned to extend the current scope to other cell lines, cells in suspension, mixed cell populations, and primary cells. Other potential extensions of HyperSeq include chromatin accessibility 12,58 , whole-genome sequencing 59 at the sequencing level, and analysis of subcellular organelles 60,61 at imaging level. Hyperseq could also be used on engineered cells that export samples of the transcriptome nondestructively 62 , potentially achieving data sets similar to Live-seq with a simpler workflow.…”

Section: Discussionmentioning

confidence: 99%

Integration of hyperspectral imaging and transcriptomics from individual cells with HyperSeq

Xie,

Habibalahi,

Anwer

et al. 2024

Preprint

View full text Add to dashboard Cite

Microscopy and omics are complementary approaches to probe the molecular state of cells in health and disease, combining granularity with scalability. While important advances have been achieved over the last decade in each area, integrating both imaging- and sequencing-based assays on the same cell has proven challenging. In this study, a new approach called HyperSeq that combines hyperspectral autofluorescence imaging with transcriptomics on the same cell is demonstrated. HyperSeq was applied to Michigan Cancer Foundation 7 (MCF-7) breast cancer cells and identified a subpopulation of cells exhibiting bright autofluorescence rings at the plasma membrane in optical channel 13 (ex = 431 nm, em = 594 nm). Correlating the presence of a ring with the gene expression in the same cell indicated that ringed cells are more likely to express hallmark genes of apoptosis and less likely to express genes associated with ATP production. Further, correlation of cell morphology with gene expression suggested that multiple members of the spliceosome were downregulated in larger MCF-7 cells. Multiple genes were evenly expressed across cell sizes but also exhibited higher usage of specific exons in larger or smaller cells. Finally, correlation between gene expression and fluorescence within the spectral range of Nicotinamide adenine dinucleotide hydrogen (NADH) provided insight into the metabolic states of MCF-7 cells. These observations provided a link between the cell's optical spectrum and its internal molecular state, demonstrating the utility of HyperSeq to study cell biology at single cell resolution by integrating spectral, morphological and transcriptomic analyses into a single, streamlined workflow.

show abstract

Section: Discussionmentioning

confidence: 99%

Integration of hyperspectral imaging and transcriptomics from individual cells with HyperSeq

Xie,

Habibalahi,

Anwer

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Zhu et al [175] have created cell-based methods with a pseudo-label assignment to discover protein subcellular localization results across distinct cells with heterogeneity among single cells. Husain et al [140] presents the Hybrid subCellular Protein Localiser (HCPL) that robustly localizes single-cell subcellular protein patterns. Wang's work with mass spectrometry (MS)-based spatial proteomics [176] shows the possibility of larger dimensional feature maps and higher learning ability of computational models.…”

Section: Future Directionsmentioning

confidence: 99%

A Review for Artificial Intelligence Based Protein Subcellular Localization

Xiao,

Zou,

Wang

et al. 2024

Biomolecules

View full text Add to dashboard Cite

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer’s disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcellular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

show abstract

“…These do not provide insight about organellar morphology and depend on tedious organelle fractionation, which is limited to only a subset of the organelles. In addition, imaging-based approaches were developed (Kobayashi et al 2022;Spitzer et al 2023;Carpenter et al 2006;Heinrich et al 2021;Lu et al 2019;Husain et al 2023;Long, Yang, and Shen 2020). However, these are largely in non-polarized cultured cells.…”

Section: Introductionmentioning

confidence: 99%

Organellomics: AI-driven deep organellar phenotyping of human neurons

Molitor,

Krispin,

van Zuiden

et al. 2024

Preprint

View full text Add to dashboard Cite

Systematic assessment of organelle architectures in cells, known as the organellome, could provide valuable insights into cellular states and disease pathologies but remains largely uncharted. Here, we devised a novel pipeline combining self-supervised deep learning and transfer learning to generate a Neuronal Organellomics Vision Atlas (NOVA). Analyzing over 1.5 million confocal images of 24 distinct membrane-bound and membrane-less organelles in human neurons, we enable a simultaneous evaluation of all organelles. We show that organellomics allows the study of cellular phenotypes by quantifying the localization and morphological properties embodied in multiple different organelles, using a unified score. We further developed a strategy to superimpose all organelles, which represents a new realization of cellular state. The value of our approach is demonstrated by characterizing specific organellar responses of human neurons to stress, cytoplasmic mislocalization of TDP-43, or disease-associated variations in ALS genes. Therefore, organellomics offers a novel approach to study the neuro-cellular biology of diseases.

show abstract

Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures

Cited by 6 publications

References 29 publications

Integration of hyperspectral imaging and transcriptomics from individual cells with HyperSeq

Integration of hyperspectral imaging and transcriptomics from individual cells with HyperSeq

A Review for Artificial Intelligence Based Protein Subcellular Localization

Organellomics: AI-driven deep organellar phenotyping of human neurons

Contact Info

Product

Resources

About