Academic literature on the topic 'Cell Annotation'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Cell Annotation.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Journal articles on the topic "Cell Annotation"
1
Huang, Xiaoqian, Ruiqi Liu, Shiwei Yang, Xiaozhou Chen, and Huamei Li. "scAnnoX: an R package integrating multiple public tools for single-cell annotation." PeerJ 12 (March 28, 2024): e17184. http://dx.doi.org/10.7717/peerj.17184.
Full textAbstract:
Background Single-cell annotation plays a crucial role in the analysis of single-cell genomics data. Despite the existence of numerous single-cell annotation algorithms, a comprehensive tool for integrating and comparing these algorithms is also lacking. Methods This study meticulously investigated a plethora of widely adopted single-cell annotation algorithms. Ten single-cell annotation algorithms were selected based on the classification of either reference dataset-dependent or marker gene-dependent approaches. These algorithms included SingleR, Seurat, sciBet, scmap, CHETAH, scSorter, sc.type, cellID, scCATCH, and SCINA. Building upon these algorithms, we developed an R package named scAnnoX for the integration and comparative analysis of single-cell annotation algorithms. Results The development of the scAnnoX software package provides a cohesive framework for annotating cells in scRNA-seq data, enabling researchers to more efficiently perform comparative analyses among the cell type annotations contained in scRNA-seq datasets. The integrated environment of scAnnoX streamlines the testing, evaluation, and comparison processes among various algorithms. Among the ten annotation tools evaluated, SingleR, Seurat, sciBet, and scSorter emerged as top-performing algorithms in terms of prediction accuracy, with SingleR and sciBet demonstrating particularly superior performance, offering guidance for users. Interested parties can access the scAnnoX package at https://github.com/XQ-hub/scAnnoX.
2
Vădineanu, Serban, Daniël M. Pelt, Oleh Dzyubachyk, and Kees Joost Batenburg. "Reducing Manual Annotation Costs for Cell Segmentation by Upgrading Low-Quality Annotations." Journal of Imaging 10, no. 7 (July 17, 2024): 172. http://dx.doi.org/10.3390/jimaging10070172.
Full textAbstract:
Deep-learning algorithms for cell segmentation typically require large data sets with high-quality annotations to be trained with. However, the annotation cost for obtaining such sets may prove to be prohibitively expensive. Our work aims to reduce the time necessary to create high-quality annotations of cell images by using a relatively small well-annotated data set for training a convolutional neural network to upgrade lower-quality annotations, produced at lower annotation costs. We investigate the performance of our solution when upgrading the annotation quality for labels affected by three types of annotation error: omission, inclusion, and bias. We observe that our method can upgrade annotations affected by high error levels from 0.3 to 0.9 Dice similarity with the ground-truth annotations. We also show that a relatively small well-annotated set enlarged with samples with upgraded annotations can be used to train better-performing cell segmentation networks compared to training only on the well-annotated set. Moreover, we present a use case where our solution can be successfully employed to increase the quality of the predictions of a segmentation network trained on just 10 annotated samples.
3
Hia, Nazifa Tasnim, and Sumon Ahmed. "Automatic cell type annotation using supervised classification: A systematic literature review." Systematic Literature Review and Meta-Analysis Journal 3, no. 3 (October 21, 2022): 99–108. http://dx.doi.org/10.54480/slrm.v3i3.45.
Full textAbstract:
Single-cell sequencing gives us the opportunity to analyze cells on an individual level rather than at a population level. There are different types of sequencing based on the stage and portion of the cell from where the data are collected. Among those Single Cell RNA seq is most widely used and most application of cell type annotation has been on Single-cell RNA seq data. Tools have been developed for automatic cell type annotation as manual annotation of cell type is time-consuming and partially subjective. There are mainly three strategies to associate cell type with gene expression profiles of single cell by using marker genes databases, correlating expression data, transferring levels by supervised classification. In this SLR, we present a comprehensive evaluation of the available tools and the underlying approaches to perform automated cell type annotations on scRNA-seq data.
4
Xu, Yang, Simon J. Baumgart, Christian M. Stegmann, and Sikander Hayat. "MACA: marker-based automatic cell-type annotation for single-cell expression data." Bioinformatics 38, no. 6 (December 22, 2021): 1756–60. http://dx.doi.org/10.1093/bioinformatics/btab840.
Full textAbstract:
Abstract Summary Accurately identifying cell types is a critical step in single-cell sequencing analyses. Here, we present marker-based automatic cell-type annotation (MACA), a new tool for annotating single-cell transcriptomics datasets. We developed MACA by testing four cell-type scoring methods with two public cell-marker databases as reference in six single-cell studies. MACA compares favorably to four existing marker-based cell-type annotation methods in terms of accuracy and speed. We show that MACA can annotate a large single-nuclei RNA-seq study in minutes on human hearts with ∼290K cells. MACA scales easily to large datasets and can broadly help experts to annotate cell types in single-cell transcriptomics datasets, and we envision MACA provides a new opportunity for integration and standardization of cell-type annotation across multiple datasets. Availability and implementation MACA is written in python and released under GNU General Public License v3.0. The source code is available at https://github.com/ImXman/MACA. Supplementary information Supplementary data are available at Bioinformatics online.
5
Gill, Jaidip, Abhijit Dasgupta, Brychan Manry, and Natasha Markuzon. "Abstract 4927: Combining single-cell ATAC and RNA sequencing for supervised cell annotation." Cancer Research 84, no. 6_Supplement (March 22, 2024): 4927. http://dx.doi.org/10.1158/1538-7445.am2024-4927.
Full textAbstract:
Abstract Background: Analysis of samples at the single-cell level offers insights into cellular heterogeneity and cell function. Cell type annotation is the first critical step for performing such an analysis. While current methods primarily utilize single-cell RNA sequencing (scRNA-seq) for annotation, several studies have demonstrated improved classification accuracy by combining scRNA-seq with transposase-accessible chromatin sequencing (ATAC-seq) using unsupervised methods. However, the utility of ATAC-seq features for supervised cell-type annotation has not been explored. Aims/Objectives: The objective of this study was to evaluate the relative performance of supervised cell-type classification using scRNA-seq alone vs. in multimodal combination with ATAC-seq; and how these data interplay with choice of classification and dimensionality reduction methods. Methods: A peripheral-blood mononuclear cell multi-omic dataset from a single, healthy female donor wasanalysed in this study. Ground truth annotations were generated using unsupervised annotation with the weighted nearest neighbour clustering method. Two dimensionality reduction methods (principal component analysis (PCA), single-cell Variational Inference (scVI) autoencoder) and four classification models (logistic regression, random forest, support vector machine (SVM)) were implemented and performance metrics (F1 score, precision, and recall) were compared over 10 bootstrap samples. Results: ATAC-seq features improved annotation quality and prediction confidence when using scVI embeddings, independent of the classifier. The best-performing model (SVM with scVI embeddings) showed an increase from a median macro F1 score of 0.907 (IQR = [0.902, 0.910]) using scRNA-seq alone to 0.946 (IQR = [0.940, 0.949], p <0.05) with ATAC-seq added. For PCA embeddings, improvements in macro F1 score were insignificant. All cell types (B, T, monocytes, natural killer and dendritic cells) showed significant improvements when using ATAC-seq with scVI embeddings. CD4 T effector memory cells showed the largest gain in F1 score (0.112, p <0.01), whilst type-2 conventional dendritic cells showed the smallest improvement (0.006, p <0.05). Prediction confidence was improved in B cells, monocytes, natural killer cells, CD4 and CD8 naïve cells, CD4 T central memory cells and CD8 T effector memory cells. Improvements in F1 scores were lost when only classifying major cell types rather than subtypes. Conclusions: Employing ATAC-seq embeddings with scVI autoencoder enhances supervised annotation quality over scRNA-only methods. Further studies should explore the use of ATAC to improve the annotation of highly heterogeneous tissues such as tumours. Citation Format: Jaidip Gill, Abhijit Dasgupta, Brychan Manry, Natasha Markuzon. Combining single-cell ATAC and RNA sequencing for supervised cell annotation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4927.
6
Zhou, Xiao, Miao Gu, and Zhen Cheng. "Local Integral Regression Network for Cell Nuclei Detection." Entropy 23, no. 10 (October 14, 2021): 1336. http://dx.doi.org/10.3390/e23101336.
Full textAbstract:
Nuclei detection is a fundamental task in the field of histopathology image analysis and remains challenging due to cellular heterogeneity. Recent studies explore convolutional neural networks to either isolate them with sophisticated boundaries (segmentation-based methods) or locate the centroids of the nuclei (counting-based approaches). Although these two methods have demonstrated superior success, their fully supervised training demands considerable and laborious pixel-wise annotations manually labeled by pathology experts. To alleviate such tedious effort and reduce the annotation cost, we propose a novel local integral regression network (LIRNet) that allows both fully and weakly supervised learning (FSL/WSL) frameworks for nuclei detection. Furthermore, the LIRNet can output an exquisite density map of nuclei, in which the localization of each nucleus is barely affected by the post-processing algorithms. The quantitative experimental results demonstrate that the FSL version of the LIRNet achieves a state-of-the-art performance compared to other counterparts. In addition, the WSL version has exhibited a competitive detection performance and an effortless data annotation that requires only 17.5% of the annotation effort.
7
Zhou, Xiao, Miao Gu, and Zhen Cheng. "Local Integral Regression Network for Cell Nuclei Detection." Entropy 23, no. 10 (October 14, 2021): 1336. http://dx.doi.org/10.3390/e23101336.
Full textAbstract:
Nuclei detection is a fundamental task in the field of histopathology image analysis and remains challenging due to cellular heterogeneity. Recent studies explore convolutional neural networks to either isolate them with sophisticated boundaries (segmentation-based methods) or locate the centroids of the nuclei (counting-based approaches). Although these two methods have demonstrated superior success, their fully supervised training demands considerable and laborious pixel-wise annotations manually labeled by pathology experts. To alleviate such tedious effort and reduce the annotation cost, we propose a novel local integral regression network (LIRNet) that allows both fully and weakly supervised learning (FSL/WSL) frameworks for nuclei detection. Furthermore, the LIRNet can output an exquisite density map of nuclei, in which the localization of each nucleus is barely affected by the post-processing algorithms. The quantitative experimental results demonstrate that the FSL version of the LIRNet achieves a state-of-the-art performance compared to other counterparts. In addition, the WSL version has exhibited a competitive detection performance and an effortless data annotation that requires only 17.5% of the annotation effort.
8
Cheng, Changde, Wenan Chen, Hongjian Jin, and Xiang Chen. "A Review of Single-Cell RNA-Seq Annotation, Integration, and Cell–Cell Communication." Cells 12, no. 15 (July 30, 2023): 1970. http://dx.doi.org/10.3390/cells12151970.
Full textAbstract:
Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool for investigating cellular biology at an unprecedented resolution, enabling the characterization of cellular heterogeneity, identification of rare but significant cell types, and exploration of cell–cell communications and interactions. Its broad applications span both basic and clinical research domains. In this comprehensive review, we survey the current landscape of scRNA-seq analysis methods and tools, focusing on count modeling, cell-type annotation, data integration, including spatial transcriptomics, and the inference of cell–cell communication. We review the challenges encountered in scRNA-seq analysis, including issues of sparsity or low expression, reliability of cell annotation, and assumptions in data integration, and discuss the potential impact of suboptimal clustering and differential expression analysis tools on downstream analyses, particularly in identifying cell subpopulations. Finally, we discuss recent advancements and future directions for enhancing scRNA-seq analysis. Specifically, we highlight the development of novel tools for annotating single-cell data, integrating and interpreting multimodal datasets covering transcriptomics, epigenomics, and proteomics, and inferring cellular communication networks. By elucidating the latest progress and innovation, we provide a comprehensive overview of the rapidly advancing field of scRNA-seq analysis.
9
Long, Helen, Richard Reeves, and Michelle M. Simon. "Mouse genomic and cellular annotations." Mammalian Genome 33, no. 1 (February 5, 2022): 19–30. http://dx.doi.org/10.1007/s00335-021-09936-7.
Full textAbstract:
AbstractMice have emerged as one of the most popular and valuable model organisms in the research of human biology. This is due to their genetic and physiological similarity to humans, short generation times, availability of genetically homologous inbred strains, and relatively easy laboratory maintenance. Therefore, following the release of the initial human reference genome, the generation of the mouse reference genome was prioritised and represented an important scientific resource for the mouse genetics community. In 2002, the Mouse Genome Sequencing Consortium published an initial draft of the mouse reference genome which contained ~ 96% of the euchromatic genome of female C57BL/6 J mice. Almost two decades on from the publication of the initial draft, sequencing efforts have continued to increase the completeness and accuracy of the C57BL/6 J reference genome alongside advances in genome annotation. Additionally new sequencing technologies have provided a wealth of data that has added to the repertoire of annotations associated with traditional genomic annotations. Including but not limited to advances in regulatory elements, the 3D genome and individual cellular states. In this review we focus on the reference genome C57BL/6 J and summarise the different aspects of genomic and cellular annotations, as well as their relevance to mouse genetic research. We denote a genomic annotation as a functional unit of the genome. Cellular annotations are annotations of cell type or state, defined by the transcriptomic expression profile of a cell. Due to the wide-ranging number and diversity of annotations describing the mouse genome, we focus on gene, repeat and regulatory element annotation as well as two relatively new technologies; 3D genome architecture and single-cell sequencing outlining their utility in genetic research and their current challenges.
10
Wei, Ziyang, and Shuqin Zhang. "CALLR: a semi-supervised cell-type annotation method for single-cell RNA sequencing data." Bioinformatics 37, Supplement_1 (July 1, 2021): i51—i58. http://dx.doi.org/10.1093/bioinformatics/btab286.
Full textAbstract:
Abstract Motivation Single-cell RNA sequencing (scRNA-seq) technology has been widely applied to capture the heterogeneity of different cell types within complex tissues. An essential step in scRNA-seq data analysis is the annotation of cell types. Traditional cell-type annotation is mainly clustering the cells first, and then using the aggregated cluster-level expression profiles and the marker genes to label each cluster. Such methods are greatly dependent on the clustering results, which are insufficient for accurate annotation. Results In this article, we propose a semi-supervised learning method for cell-type annotation called CALLR. It combines unsupervised learning represented by the graph Laplacian matrix constructed from all the cells and supervised learning using sparse logistic regression. By alternately updating the cell clusters and annotation labels, high annotation accuracy can be achieved. The model is formulated as an optimization problem, and a computationally efficient algorithm is developed to solve it. Experiments on 10 real datasets show that CALLR outperforms the compared (semi-)supervised learning methods, and the popular clustering methods. Availability and implementation The implementation of CALLR is available at https://github.com/MathSZhang/CALLR. Supplementary information Supplementary data are available at Bioinformatics online.
Dissertations / Theses on the topic "Cell Annotation"
1
Raoux, Corentin. "Review and Analysis of single-cell RNA sequencing cell-type identification and annotation tools." Thesis, KTH, Skolan för kemi, bioteknologi och hälsa (CBH), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-297852.
Full textAbstract:
Single-cell RNA-sequencing makes possible to study the gene expression at the level of individual cells. However, one of the main challenges of the single-cell RNA-sequencing analysis today, is the identification and annotation of cell types. The current method consists in manually checking the expression of genes using top differentially expressed genes and comparing them with related cell-type markers available in scientific publications. It is therefore time-consuming and labour intensive. Nevertheless, in the last two years,numerous automatic cell-type identification and annotation tools which use different strategies have been created. But, the lack of specific comparisons of those tools in the literature and especially for immuno-oncologic and oncologic purposes makes difficult for laboratories and companies to know objectively what are the best tools for annotating cell types. In this project, a review of the current tools and an evaluation of R tools were carried out.The annotation performance, the computation time and the ease of use were assessed. After this preliminary results, the best selected R tools seem to be ClustifyR (fast and rather precise) and SingleR (precise) for the correlation-based tools, and SingleCellNet (precise and rather fast) and scPred (precise but a lot of cell types remains unassigned) for the supervised classificationtools. Finally, for the marker-based tools, MAESTRO and SCINA are rather robust if they are provided with high quality markers.
2
Ebenezer, ThankGod Echezona. "The genome of Euglena gracilis : annotation, function and expression." Thesis, University of Cambridge, 2018. https://www.repository.cam.ac.uk/handle/1810/275885.
Full textAbstract:
Euglena gracilis is a species of unicellular photosynthetic flagellate that inhibits aquatic ecosystems. E. gracilis belongs to the supergroup Excavata, and are an important component of the global biosphere, have biotechnological potential and is useful biological model due to their evolutionary history and complex biology. Whilst the evolutionary position of E. gracilis is now clear, their relationship with other protists such as Naegleria, Giardia, and Kinetoplastids, remains to be investigated in detail. Investigating and understanding the biology of this complex organism is a promising way to approach many evolutionary puzzles, including secondary endosymbiotic events and the evolution of parasitism, due to their relationship with Kinetoplastids. Here, I report a draft genome for E. gracilis, together with a high quality transcriptome and proteomic analysis. The estimated genome size is ~ 2 Gbp, with a GC content of ~ 50 % and a protein coding potential predicted at 36,526 Open Reading Frames (ORFs). Less than 25% of the genome is single copy sequence, indicating extensive repeat structure. There are evidences for large number of paralogs amongst specific gene families, indicating expansions and possible polyploidy as well as extensive sharing of genes with other non photosynthetic and photosynthetic eukaryotes: red and green algael genes, together with trypanosomes and other members of the excavates. Functional resolution into several of the biological systems indicates multiple similarities with the trypanosomatids in terms of orthology, paralogy, relatedness and complexity. Several biological systems such as nuclear architecture (e.g. chromosome segregation, nuclear pore complex, nuclear lamins), protein trafficking, translation, surface, consist of conserved and divergent components. For instance, several gene families likely associated with the cell surface and signal transduction possess very large numbers of lineage-specific paralogs, suggesting great flexibility in environmental monitoring and, together with divergent mechanisms for metabolic control, novel solutions to adaptation to extreme environments. I also demonstrate that the majority of control of protein expression levels is post-transcriptional and absence of transcriptional regulation, despite the presence of conventional introns. These data are a major advance in the understanding of the nuclear genome of Euglenids and provide a platform for investigation of the contributions of E. gracilis and relatives to the biosphere.
3
Abbey, Simon. "Annotation of the human odontoblast cell layer and dental pulp proteomes and N-terminomes." Thesis, University of British Columbia, 2017. http://hdl.handle.net/2429/62403.
Full textAbstract:
The proteome and N-terminome of the human odontoblast cell layer was identified for the first time by shotgun proteomic and terminal amine isotopic labeling of substrates (TAILS) N terminomic analyses, respectively, and compared with that of human dental pulp stroma from 3rd molar teeth. After reverse-phase liquid chromatography-tandem mass spectrometry, >170,000 spectra from the shotgun and TAILS analyses were matched by four search engines to 4,888 and 12,063 peptides in the odontoblast cell layer and pulp stroma, respectively. Using the Trans-Proteomic Pipeline, I identified 895 and 2,423 unique proteins in these tissues at an FDR of ≤ 1 %. In the odontoblast cell layer proteome I found proteomic evidence for dentin sialophosphoprotein, which is cleaved into dentin phosphoprotein and dentin sialoprotein, proteins that are important in dentin mineralization. Further, 222 proteins of the odontoblast cell layer were not found in the pulp, suggesting many of these proteins are synthesized preferentially by odontoblasts. I also found minor differences in the odontoblast cell layer between the dental pulp proteomes of older and younger donors. The human dental pulp stroma proteome was expanded by 974 new proteins, not previously identified among the 4,123 proteins identified in our previous dental pulp study (Eckhard et al., 2015). Thus, by exploring the proteome of the odontoblast cell layer and expanding the known dental pulp proteome, we found distinct proteome differences when compared with each other and with dentin. The mass spectrometry raw data and metadata have been deposited to ProteomeXchange with the PXD identifier ˂PXD006557˃.Dentistry, Faculty of
Graduate
4
Hoffman, Matthew P. "The Cortical response to RhoA is regulated during mitosis. Annotation of cytoskeletal and motility proteins in the sea urchin genome assembly." Thesis, Boston College, 2008. http://hdl.handle.net/2345/bc-ir:107671.
Full textAbstract:
Thesis advisor: David BurgessThis doctoral thesis addresses two central topics divided into separate chapters. In Chapter 1: The cortical response to RhoA is regulated during mitosis, experimental findings using sea urchin embryos are presented that demonstrate that the small GTPase RhoA participates in positive signaling for cell division and that this activity is negatively regulated prior to anaphase. In a second series of experiments, myosin phosphatase is shown to be a central negative regulator of myosin activity during the cell cycle through metaphase of mitosis and experimental findings support the conclusion that myosin phosphatase opposes RhoA signaling until anaphase onset. These experiments also reveal that myosin activation alone is insufficient to stimulate cortical contractions during S phase and during metaphase arrest following activation of the spindle checkpoint. In Chapter 2: Annotation of cytoskeletal and motility proteins in the sea urchin genome assembly, as part of a collaborative project, homologs of cytoskeletal genes and gene families were derived and annotated from the sea urchin genome assembly. In addition, phylogenetic analysis of multiple gene families is presented based on these findings
Thesis (PhD) — Boston College, 2008
Submitted to: Boston College. Graduate School of Arts and Sciences
Discipline: Biology
5
Collin, Antoine. "Annotation cellulaire automatique pour la construction d'un atlas cellulaire." Electronic Thesis or Diss., Université Côte d'Azur, 2024. http://www.theses.fr/2024COAZ6039.
Full textAbstract:
Les technologies d'analyse de l'expression des gènes sur cellule unique, apparues depuis une dizaine d'années, sont en train de modifier profondément les approches de biologie cellulaire.L'analyse de données de cellules uniques est un processus complexe. Une étape clé est l'annotation cellulaire, qui consiste à assigner un type cellulaire pertinent à chacune des cellules analysées. La bonne annotation des cellules conditionne la qualité des analyses ultérieures. Cette tâche réclame une expertise biologique sur le tissu d'intérêt et une expertise computionnelle en analyse de données. Des initiatives tel que le Human Cell atlas (HCA) permettent de disposer d'atlas de référence de grande taille dotées d'annotations méticuleuses. En tant que telles, elles constituent une opportunité pour développer des modèles d'apprentissage automatique profonds susceptibles d'automatiser le processus d'annotation.Les enjeux de cette thèse étaient de mettre en place des outils d'annotation cellulaire automatiques pouvant fonctionner sur de grands jeux de données. Pour y parvenir, deux axes de travail ont été développés: j'ai tout d'abord réalisé l'annotation d'un atlas des voies aériennes humaines comprenant plus de 400.000 cellules à partir de plusieurs dizaines de biopsies obtenues chez des patients atteints de formes précoces de bronchopneumopathie chronique obstructive (BPCO), qui ont été comparées à autant de biopsies provenant de volontaires sains de même âge. J'ai ensuite mis au point une méthode d'annotation automatique après avoir réalisé un état de l'art des outils existants.Dans une première partie, mon analyse bioinformatique a permis de caractériser le rôle central joué par la fumée de cigarette, principalement au niveau des cellules épithéliales situées à la surface des voies aériennes trachéobronchiques, et dès lors directement en contact avec la fumée de cigarette. Les populations cellulaires affectées sont caractérisées par l'expression de gènes codant pour des enzymes de détoxification ou impliquées dans le métabolisme xénobiotique, l'expression d'aucun de ces gènes n'étant affectée chez d'anciens fumeurs, ni chez des volontaires sains. Cette réversibilité phénotypique lors de l'arrêt de consommation du tabac s'accompagne de modifications moléculaires et cliniques liés à la BPCO. Mes travaux sont actuellement complétés par des approches de transcriptomique spatiale, et d'analyse de l'expression des différentes isoformes de transcrits.La seconde partie de cette thèse explore les méthodes d'annotation automatiques existantes à la lumière des problématiques rencontrées lors de l'annotation de l'atlas BPCO. J'ai d'abord effectué une revue extensive de la littérature, avec un intérêt particulier pour les méthodes utilisant des modèles d'apprentissage profond. J'ai ensuite développé mon propre outil d'annotation automatique, scMusketeers, dont l'architecture favorise la construction d'un espace latent renforçant le type cellulaire tout en minimisant les effets inter-batchs expérimentaux. Des tests menés sur 12 jeux de données différents sur 7 outils actuellement disponibles le positionne favorablement, notamment pour la détection de types cellulaires raresSingle-cell gene expression analysis technologies, which have emerged over the last ten years, are profoundly changing approaches to cell biology. The analysis of single-cell data is a complex process involving many steps. A key step is cell annotation, which involves assigning the most relevant cell type to the different cells analysed. Correct cell annotation determines the quality of subsequent analyses. This complex task requires biological expertise of the tissue of interest and computational expertise to carry out data analysis. Initiatives such as the HCA provide large reference atlases with curated annotation. As such, they represent an opportunity to develop deep learning models capable of automating the annotation process.The aim of this thesis was to set up automatic annotation tools that could operate on large datasets. To achieve this, two lines of work were developed: first, I created an atlas of the human airways comprising more than 400,000 cells based on several dozen biopsies obtained from patients with early forms of chronic obstructive pulmonary disease (COPD), which were compared with as many biopsies from healthy volunteers of the same age. I then developed an automatic annotation method after reviewing the state of the art existing tools.In the first part, I was able to characterise the central role played by cigarette smoke, mainly in the epithelial cells located on the surface of the tracheobronchial airways, and thus directly exposed to cigarette smoke. The populations affected are characterised by the expression of genes coding for detoxification enzymes or enzymes involved in xenobiotic metabolism. None of these genes were affected in either ex-smokers or healthy patients. There appears to be a reversibility of the pathology following cessation of smoking, despite the molecular changes induced during initial exposure to cigarette smoke. My work is now currently extended by spatial transcriptomic approaches and analysis of the expression of different transcript isoforms.In the second part, I explored existing automatic annotation methods in the light of the knowledge acquired when annotating the COPD atlas. I began with an extensive review of the literature, with a particular interest in methods using deep learning models. I then developed an automatic annotation tool, scMusketeers, whose architecture favours the construction of a latent space reinforcing cell type while minimising experimental inter-batch effects. It compared favorably to 7 currently available tools on 12 different datasets, particularly in the task of detecting rare cell types
6
Lehmann, Nathalie. "Development of bioinformatics tools for single-cell transcriptomics applied to the search for signatures of symmetric versus asymmetric division mode in neural progenitors." Electronic Thesis or Diss., Université Paris sciences et lettres, 2021. http://www.theses.fr/2021UPSLE070.
Full textAbstract:
Ces dernières années, l’émergence des approches en cellules uniques (scRNA-seq) a favorisé la caractérisation de l’hétérogénéité cellulaire avec une précision inégalée. Malgré leur démocratisation, l’analyse de ces données reste complexe, en particulier pour les organismes dont les annotations sont incomplètes. Au cours ma thèse, j’ai observé que les annotations génomiques du poulet sont lacunaires, ce qui engendre la perte d’un grand nombre de lectures de séquençage. J’ai évalué à quel point une annotation améliorée affecte les résultats biologiques et les conclusions issues de ces analyses. Nous proposons une nouvelle approche basée sur la ré-annotation du génome à partir de données scRNA-seq et de RNA-seq bulk en lectures longues. Ce projet de biologie computationnelle s’appuie sur une étroite collaboration avec l’équipe expérimentale de Xavier Morin (IBENS). Le principal objectif biologique est la recherche de signatures de mode de division symétrique et asymétrique au sein de progéniteurs neuronaux. Afin d’identifier les principaux changements transcriptionnels, j’ai mis en place des approches dédiées à la recherche de signatures géniques à partir de données scRNA-seqIn recent years, single-cell RNA-seq (scRNA-seq) has fostered the characterization of cell heterogeneity at a remarkable high resolution. Despite their democratization, the analysis of scRNA-seq remains a challenge, particularly for organisms whose genomic annotations are partial. During my PhD, I observed that the chick genomic annotations are often incomplete, thus resulting in a loss of a large number of sequencing reads. I investigated how an enriched annotation affects the biological results and conclusions from these analyses. We developed a novel approach based on the re-annotation of the genome with scRNA-seq data and long reads bulk RNA-seq. This computational biology project capitalises on a tight collaboration with the experimental team of Xavier Morin (IBENS). The main biological focus is the search for signatures of symmetric versus asymmetric division mode in neural progenitors. In order to identify the key transcriptional switches that occur during the neurogenic transition, I have implemented bioanalysis approaches dedicated to the search for gene signatures from scRNA-seq data
7
Planas, Iglesias Joan 1980. "On the study of 3D structure of proteins for developing new algorithms to complete the interactome and cell signalling networks." Doctoral thesis, Universitat Pompeu Fabra, 2013. http://hdl.handle.net/10803/104152.
Full textAbstract:
Proteins are indispensable players in virtually all biological events. The functions of proteins are determined by their three dimensional (3D) structure and coordinated through intricate networks of protein-protein interactions (PPIs). Hence, a deep comprehension of such networks turns out to be crucial for understanding the cellular biology. Computational approaches have become critical tools for analysing PPI networks. In silico methods take advantage of the existing PPI knowledge to both predict new interactions and predict the function of proteins. Regarding the task of predicting PPIs, several methods have been already developed. However, recent findings demonstrate that such methods could take advantage of the knowledge on non-interacting protein pairs (NIPs). On the task of predicting the function of proteins,the Guilt-by-Association (GBA) principle can be exploited to extend the functional annotation of proteins over PPI networks. In this thesis, a new algorithm for PPI prediction and a protocol to complete cell signalling networks are presented. iLoops is a method that uses NIP data and structural information of proteins to predict the binding fate of protein pairs. A novel protocol for completing signalling networks –a task related to predicting the function of a protein, has also been developed. The protocol is based on the application of GBA principle in PPI networks.Les proteïnes tenen un paper indispensable en virtualment qualsevol procés biològic. Les funcions de les proteïnes estan determinades per la seva estructura tridimensional (3D) i són coordinades per mitjà d’una complexa xarxa d’interaccions protiques (en anglès, protein-protein interactions, PPIs). Axí doncs, una comprensió en profunditat d’aquestes xarxes és fonamental per entendre la biologia cel•lular. Per a l’anàlisi de les xarxes d’interacció de proteïnes, l’ús de tècniques computacionals ha esdevingut fonamental als darrers temps. Els mètodes in silico aprofiten el coneixement actual sobre les interaccions proteiques per fer prediccions de noves interaccions o de les funcions de les proteïnes. Actualment existeixen diferents mètodes per a la predicció de noves interaccions de proteines. De tota manera, resultats recents demostren que aquests mètodes poden beneficiar-se del coneixement sobre parelles de proteïnes no interaccionants (en anglès, non-interacting pairs, NIPs). Per a la tasca de predir la funció de les proteïnes, el principi de “culpable per associació” (en anglès, guilt by association, GBA) és usat per extendre l’anotació de proteïnes de funció coneguda a través de xarxes d’interacció de proteïnes. En aquesta tesi es presenta un nou mètode pre a la predicció d’interaccions proteiques i un nou protocol basat per a completar xarxes de senyalització cel•lular. iLoops és un mètode que utilitza dades de parells no interaccionants i coneixement de l’estructura 3D de les proteïnes per a predir interaccions de proteïnes. També s’ha desenvolupat un nou protocol per a completar xarxes de senyalització cel•lular, una tasca relacionada amb la predicció de les funcions de les proteïnes. Aquest protocol es basa en aplicar el principi GBA a xarxes d’interaccions proteiques.
8
Ghadie, Mohamed A. "Analysis and Reconstruction of the Hematopoietic Stem Cell Differentiation Tree: A Linear Programming Approach for Gene Selection." Thesis, Université d'Ottawa / University of Ottawa, 2015. http://hdl.handle.net/10393/32048.
Full textAbstract:
Stem cells differentiate through an organized hierarchy of intermediate cell types to terminally differentiated cell types. This process is largely guided by master transcriptional regulators, but it also depends on the expression of many other types of genes. The discrete cell types in the differentiation hierarchy are often identified based on the expression or non-expression of certain marker genes. Historically, these have often been various cell-surface proteins, which are fairly easy to assay biochemically but are not necessarily causative of the cell type, in the sense of being master transcriptional regulators. This raises important questions about how gene expression across the whole genome controls or reflects cell state, and in particular, differentiation hierarchies. Traditional approaches to understanding gene expression patterns across multiple conditions, such as principal components analysis or K-means clustering, can group cell types based on gene expression, but they do so without knowledge of the differentiation hierarchy. Hierarchical clustering and maximization of parsimony can organize the cell types into a tree, but in general this tree is different from the differentiation hierarchy. Using hematopoietic differentiation as an example, we demonstrate how many genes other than marker genes are able to discriminate between different branches of the differentiation tree by proposing two models for detecting genes that are up-regulated or down-regulated in distinct lineages. We then propose a novel approach to solving the following problem: Given the differentiation hierarchy and gene expression data at each node, construct a weighted Euclidean distance metric such that the minimum spanning tree with respect to that metric is precisely the given differentiation hierarchy. We provide a set of linear constraints that are provably sufficient for the desired construction and a linear programming framework to identify sparse sets of weights, effectively identifying genes that are most relevant for discriminating different parts of the tree. We apply our method to microarray gene expression data describing 38 cell types in the hematopoiesis hierarchy, constructing a sparse weighted Euclidean metric that uses just 175 genes. These 175 genes are different than the marker genes that were used to identify the 38 cell types, hence offering a novel alternative way of discriminating different branches of the tree. A DAVID functional annotation analysis shows that the 175 genes reflect major processes and pathways active in different parts of the tree. However, we find that there are many alternative sets of weights that satisfy the linear constraints. Thus, in the style of random-forest training, we also construct metrics based on random subsets of the genes and compare them to the metric of 175 genes. Our results show that the 175 genes frequently appear in the random metrics, implicating their significance from an empirical point of view as well. Finally, we show how our linear programming method is able to identify columns that were selected to build minimum spanning trees on the nodes of random variable-size matrices.
9
Khattra, Jaswinder. "Cloning and annotation of novel transcripts from human embryonic stem cells." Thesis, University of British Columbia, 2007. http://hdl.handle.net/2429/343.
Full textAbstract:
Both cDNA tag-based and DNA chip hybridization assays have revealed widespread transcriptional activity across mammalian genomes, providing a rich source of novel protein-coding and non-coding transcripts. Annotation and functional evaluation of this undefined transcriptome space represents a major step towards the comprehensive definition of biomolecules regulating the properties of living cells, including embryonic stem cells (ESCs) and their derivatives.
In this study I analysed 87 rare mRNA transcripts from human ESCs that mapped uniquely to the human genome, in regions lacking evidence for known genes or transcripts. In addition, the transcripts appeared enriched in the hESC transcriptome as enumerated by serial analysis of gene expression (SAGE). Full-length transcripts corresponding to twelve novel LongSAGE tags were recovered and evaluated with respect to gene structure, protein-coding potential, and gene regulatory features. In addition, transcript abundance was compared between RNA isolated from undifferentiated hESCs and differentiated cells.
Analysis of full-length transcripts revealed that the novel ORFs did not exceed a size of129 amino acids and no matches were observed to well characterized protein domains. Interesting protein level predictions included small disulfide-bonded proteins, known members of which are important in a variety of biological processes. Transcripts evaluated for differential expression by real-time RT-qPCR (Reverse Transcription followed by real-time quantitative Polymerase Chain Reaction) were found to be variably expressed (0.2- to 4.5-fold) in Day-2 orDay-4 retinoic acid-induced differentiation cultures compared to undifferentiated hESCs.
Relative quantitation using a universal reference RNA (derived from pooled adult tissues)showed large differences in novel transcript levels (0.002- to 35-fold) compared to hESCs.
Collectively, these results provide a detailed analysis of a set of novel hESC transcripts and their abundance in early and adult differentiated cell types, both of which may advance our understanding of the transcriptional events governing stem cell behavior.
10
Lux, Markus [Verfasser], and Barbara [Akademischer Betreuer] Hammer. "Efficient Grouping Methods for the Annotation and Sorting of Single Cells / Markus Lux ; Betreuer: Barbara Hammer." Bielefeld : Universitätsbibliothek Bielefeld, 2018. http://d-nb.info/1160033226/34.
Full textBook chapters on the topic "Cell Annotation"
1
Wang, Zuhui, and Zhaozheng Yin. "Annotation-Efficient Cell Counting." In Medical Image Computing and Computer Assisted Intervention – MICCAI 2021, 405–14. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-87237-3_39.
Full text
2
O’Connor, Maria F., Arthur Hughes, Chaoxin Zheng, Anthony Davies, Dermot Kelleher, and Khurshid Ahmad. "Annotation and Retrieval of Cell Images." In Intelligent Data Engineering and Automated Learning – IDEAL 2010, 218–25. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010. http://dx.doi.org/10.1007/978-3-642-15381-5_27.
Full text
3
Li, Dongshunyi, Jun Ding, and Ziv Bar-Joseph. "Unsupervised Cell Functional Annotation for Single-Cell RNA-Seq." In Lecture Notes in Computer Science, 349–52. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-031-04749-7_24.
Full text
4
Vădineanu, Şerban, Daniël M. Pelt, Oleh Dzyubachyk, and K. Joost Batenburg. "Reducing Manual Annotation Costs for Cell Segmentation by Upgrading Low-Quality Annotations." In Medical Image Learning with Limited and Noisy Data, 3–13. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-44917-8_1.
Full text
5
Valkiers, Sebastiaan, Sofie Gielis, Vincent M. L. Van Deuren, Kris Laukens, and Pieter Meysman. "Clustering and Annotation of T Cell Receptor Repertoires." In Computational Vaccine Design, 33–51. New York, NY: Springer US, 2023. http://dx.doi.org/10.1007/978-1-0716-3239-0_3.
Full text
6
Shui, Zhongyi, Shichuan Zhang, Chenglu Zhu, Bingchuan Wang, Pingyi Chen, Sunyi Zheng, and Lin Yang. "End-to-End Cell Recognition by Point Annotation." In Lecture Notes in Computer Science, 109–18. Cham: Springer Nature Switzerland, 2022. http://dx.doi.org/10.1007/978-3-031-16440-8_11.
Full text
7
Li, Chao-Ting, Hung-Wen Tsai, Tseng-Lung Yang, Jung-Chi Lin, Nan-Haw Chow, Yu Hen Hu, Kuo-Sheng Cheng, and Pau-Choo Chung. "Imbalance-Effective Active Learning in Nucleus, Lymphocyte and Plasma Cell Detection." In Interpretable and Annotation-Efficient Learning for Medical Image Computing, 223–32. Cham: Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-61166-8_24.
Full text
8
Çiçek, Özgün, Yassine Marrakchi, Enoch Boasiako Antwi, Barbara Di Ventura, and Thomas Brox. "Recovering the Imperfect: Cell Segmentation in the Presence of Dynamically Localized Proteins." In Interpretable and Annotation-Efficient Learning for Medical Image Computing, 85–93. Cham: Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-61166-8_9.
Full text
9
Bashir, Raja Muhammad Saad, Talha Qaiser, Shan E. Ahmed Raza, and Nasir M. Rajpoot. "HydraMix-Net: A Deep Multi-task Semi-supervised Learning Approach for Cell Detection and Classification." In Interpretable and Annotation-Efficient Learning for Medical Image Computing, 164–71. Cham: Springer International Publishing, 2020. http://dx.doi.org/10.1007/978-3-030-61166-8_18.
Full text
10
Khalid, Nabeel, Tiago Comassetto Froes, Maria Caroprese, Gillian Lovell, Johan Trygg, Andreas Dengel, and Sheraz Ahmed. "PACE: Point Annotation-Based Cell Segmentation for Efficient Microscopic Image Analysis." In Artificial Neural Networks and Machine Learning – ICANN 2023, 545–57. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-44210-0_44.
Full textConference papers on the topic "Cell Annotation"
1
Li, Tianhao, Yugui Xu, Sihan He, Yuhang Liu, Zixuan Wang, Zhigan Zhou, Yongqing Zhang, and Quan Zou. "Cell-Specific Highly Correlated Network for Self-Supervised Distillation in Cell Type Annotation." In 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 988–93. IEEE, 2024. https://doi.org/10.1109/bibm62325.2024.10822095.
Full text
2
Tang, Binhua, and Guowei Cheng. "A Novel GCN-Based Cell Annotation Method for Single-Cell RNA Sequencing Data." In 2024 17th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 1–5. IEEE, 2024. https://doi.org/10.1109/cisp-bmei64163.2024.10906231.
Full text
3
Li, Jiawei, Shizhan Chen, Zongbo Han, Wei Li, Jijun Tang, and Fei Guo. "Multi-Task Driven Multi-Level Dynamical Fusion for Single-Cell Multi-Omics Cell Type Annotation." In 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 1009–14. IEEE, 2024. https://doi.org/10.1109/bibm62325.2024.10822524.
Full text
4
Zhu, Chi, Fengcui Qian, Yongbin Liu, Ying Yu, Chunquan Li, and Chunping Ouyang. "PEGCN: A Single-Cell Type Annotation model based on GCN with Pseudo Labels and Ensemble Learning." In 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 1414–21. IEEE, 2024. https://doi.org/10.1109/bibm62325.2024.10822733.
Full text
5
Neelapala, Satya Deepika, Soumya Jana, and Lopamudra Giri. "U-Net-Based HeLa Cell Segmentation with Zero Manual Labeling Using DBSCAN-Generated Annotations." In 2024 IEEE International Conference on E-health Networking, Application & Services (HealthCom), 1–3. IEEE, 2024. https://doi.org/10.1109/healthcom60970.2024.10880723.
Full text
6
Chen, Jian, Chengliang Wang, Xing Wu, Longrong Ran, Zailin Yang, and Yao Liu. "Weakly Supervised Segmentation of Plasma Cells in Bone Marrow via Scribble Annotations." In 2024 International Joint Conference on Neural Networks (IJCNN), 1–8. IEEE, 2024. http://dx.doi.org/10.1109/ijcnn60899.2024.10651396.
Full text
7
Zhai, Yuyao, Liang Chen, and Minghua Deng. "Realistic Cell Type Annotation and Discovery for Single-cell RNA-seq Data." In Thirty-Second International Joint Conference on Artificial Intelligence {IJCAI-23}. California: International Joint Conferences on Artificial Intelligence Organization, 2023. http://dx.doi.org/10.24963/ijcai.2023/552.
Full textAbstract:
The rapid development of single-cell RNA sequencing (scRNA-seq) technologies allows us to explore tissue heterogeneity at the cellular level. Cell type annotation plays an essential role in the substantial downstream analysis of scRNA-seq data. Existing methods usually classify the novel cell types in target data as an “unassigned” group and rarely discover the fine-grained cell type structure among them. Besides, these methods carry risks, such as susceptibility to batch effect between reference and target data, thus further compromising of inherent discrimination of target data. Considering these limitations, here we propose a new and practical task called realistic cell type annotation and discovery for scRNA-seq data. In this task, cells from seen cell types are given class labels, while cells from novel cell types are given cluster labels. To tackle this problem, we propose an end-to-end algorithm framework called scPOT from the perspective of optimal transport (OT). Specifically, we first design an OT-based prototypical representation learning paradigm to encourage both global discriminations of clusters and local consistency of cells to uncover the intrinsic structure of target data. Then we propose an unbalanced OT-based partial alignment strategy with statistical filling to detect the cells from the seen cell types across reference and target data. Notably, scPOT also introduces an easy yet effective solution to automatically estimate the overall cell type number in target data. Extensive results on our carefully designed evaluation benchmarks demonstrate the superiority of scPOT over various state-of-the-art clustering and annotation methods.
8
Zhai, Yuyao, Liang Chen, and Minghua Deng. "Distribution-Independent Cell Type Identification for Single-Cell RNA-seq Data." In Thirty-Third International Joint Conference on Artificial Intelligence {IJCAI-24}. California: International Joint Conferences on Artificial Intelligence Organization, 2024. http://dx.doi.org/10.24963/ijcai.2024/679.
Full textAbstract:
Automatic cell type annotation aims to transfer the label knowledge from label-abundant reference data to label-scarce target data, which makes encouraging progress in single-cell RNA-seq data analysis. While previous works have focused on classifying close-set cells and detecting open-set cells during testing, it is still essential to be able to classify unknown cell types as human beings. Additionally, few efforts have been devoted to addressing the challenge of common long-tail dilemma in cell type annotation data. Therefore, in this paper, we propose an innovative distribution-independent universal cell type identification framework called scDET from the perspective of autonomously equilibrated dual-consultative contrastive learning. Our model can generate fine-grained predictions for both close-set and open-set cell types in a long-tailed open-world environment. scDET consists of a contrastive-learning branch and a pseudo-labeling branch, which work collaboratively to provide interactive supervision. Specifically, the contrastive-learning branch provides reliable distribution estimation to regularize the predictions of the pseudo-labeling branch, which in turn guides itself through self-balanced knowledge transfer and a designed novel soft contrastive loss. Extensive experimental results on various evaluation datasets demonstrate the superior performance of scDET over other state-of-the-art single-cell clustering and annotation methods.
9
Jiang, Yangbo, Shuting Zhang, Jinggen Wu, Xumei Zhu, Xiao Liu, and Nenggan Zheng. "AL-Annotator: An Active Learning-based Cervical Cell Annotation System." In 2023 IEEE 29th International Conference on Parallel and Distributed Systems (ICPADS). IEEE, 2023. http://dx.doi.org/10.1109/icpads60453.2023.00079.
Full text
10
Lin, Lu, Wen Xue, Xindian Wei, Wenjun Shen, Cheng Liu, Si Wu, and Hau San Wong. "SCTrans: Multi-scale scRNA-seq Sub-vector Completion Transformer for Gene-selective Cell Type Annotation." In Thirty-Third International Joint Conference on Artificial Intelligence {IJCAI-24}. California: International Joint Conferences on Artificial Intelligence Organization, 2024. http://dx.doi.org/10.24963/ijcai.2024/658.
Full textAbstract:
Cell type annotation is pivotal to single-cell RNA sequencing data (scRNA-seq)-based biological and medical analysis, e.g., identifying biomarkers, exploring cellular heterogeneity, and understanding disease mechanisms. The previous annotation methods typically learn a nonlinear mapping to infer cell type from gene expression vectors, and thus fall short in discovering and associating salient genes with specific cell types. To address this issue, we propose a multi-scale scRNA-seq Sub-vector Completion Transformer, and our model is referred to as SCTrans. Considering that the expressiveness of gene sub-vectors is richer than that of individual genes, we perform multi-scale partitioning on gene vectors followed by masked sub-vector completion, conditioned on unmasked ones. Toward this end, the multi-scale sub-vectors are tokenized, and the intrinsic contextual relationships are modeled via self-attention computation and conditional contrastive regularization imposed on an encoding transformer. By performing mutual learning between the encoder and an additional lightweight counterpart, the salient tokens can be distinguished from the others. As a result, we can perform gene-selective cell type annotation, which contributes to our superior performance over state-of-the-art annotation methods.
Reports on the topic "Cell Annotation"
1
Blumwald, Eduardo, and Avi Sadka. Sugar and Acid Homeostasis in Citrus Fruit. United States Department of Agriculture, January 2012. http://dx.doi.org/10.32747/2012.7697109.bard.
Full textAbstract:
Citrus fruit quality standards have been determined empirically, depending on species and on the particular growing regions. In general, the TSS (total soluble solids) to total acidity (TA) ratio determines whether citrus fruit can be marketed. Soluble sugars account for most of the TSS during harvest while TA is determined almost solely by the citric acid content, which reaches levels of 1-5% by weight in many cultivated varieties. Acid and sugar homeostasis in the fruit is critical for the management of existing cultivars, the development of new cultivars, the improvement of pre- and post-harvest strategies and the control of fruit quality and disorders. The current proposal (a continuation of a previous proposal) aimed at: (1) completing the citrus fruit proteome and metabolome, and establish a citrus fruit functional database, (2) further characterization of the control of fruit acidity by studying the regulation of key steps affecting citrate metabolism, and determine the fate of citrate during acid decline stage, and (3) Studying acid and sugar homeostasis in citrus fruits by characterizing transport mechanisms across membranes. These aims were completed as the following: (1) Our initial efforts were aimed at the characterization and identification of citric acid transporters in citrus juice cells. The identification of citrate transporters at the vacuole of the citrus juice cell indicated that the steady-state citrate cytosolic concentration and the action of the cytosolic aconitase were key elements in establishing the pH homeostat in the cell that regulates the metabolic shift towards carbon usage in the fruit during the later stages of fruit development. We focused on the action of aconitase, the enzyme mediating the metabolic use of citric acid in the cells, and identified processes that control carbon fluxes in developing citrus fruits that control the fruit acid load; (2) The regulation of aconitase, catalyzing a key step in citrate metabolism, was further characterized by using two inhibitors, citramalte and oxalomalte. These compounds significantly increased citrate content and reduced the enzyme’s activity. Metabolite profiling and changes of amino-acid metabolizing enzymes in oxalomalate- treated cells suggested that the increase in citrate, caused by aconitase inhibition, induces amino acid synthesis and the GABA shunt, in accordance with the suggested fate of citrate during the acid decline stage in citrus fruit. (3) We have placed a considerable amount of time on the development of a citrus fruit proteome that will serve to identify all of the proteins in the juice cells and will also serve as an aid to the genomics efforts of the citrus research community (validating the annotation of the fruit genes and the different ESTs). Initially, we identified more than 2,500 specific fruit proteins and were able to assign a function to more than 2,100 proteins (Katz et al., 2007). We have now developed a novel Differential Quantitative LC-MS/MS Proteomics Methodology for the identification and quantitation of key biochemical pathways in fruits (Katz et al., 2010) and applied this methodology to identify determinants of key traits for fruit quality (Katz et al., 2011). We built “biosynthesis maps” that will aid in defining key pathways associated with the development of key fruit quality traits. In addition, we constructed iCitrus (http://wiki.bioinformatics.ucdavis.edu/index.php/ICitrus), a “functional database” that is essentially a web interface to a look-up table that allows users to use functional annotations in the web to identify poorly annotated citrus proteins. This resource will serve as a tool for growers and field extension specialists.
2
Rabethge, Nico, and Kurt-Georg Ciesinger. KI in der Schmutzwäsche-Sortierung. Kompetenzzentrum Arbeitswelt.Plus, January 2023. http://dx.doi.org/10.55594/wgct6835.
Full textAbstract:
Im Leuchtturmprojekt KI in der Schmutzwäschesortierung beschäftigen wir uns mit der Entwicklung eines humanzentrierten KI-gesteuerten Identifikationsmoduls zur Sortie-rung von Wäschestücken nach Waschkategorien. Diese Entwicklung soll in der Praxis dazu beitragen, dass die hochbelastende und potenziell gesundheitsgefährdende manuelle Sortierung von Schmutzwäsche nicht mehr wie im bisherigen Umfang von Menschen geleistet werden muss. Eine KI-gestützte Identifikation, verbunden mit einem Handhabungsgerät könnte die Sortierung in weiten Teilen autonom übernehmen. Neben der optimalen Unterstützung von Mitarbeitenden sind die Erklärung und Transferierbarkeit der KI, sowie die Entwicklung eines Schulungsprogramms, um Akzeptanz sowie Verständnis bei den Nutzer:innen zu verstärken, Forschungsschwerpunkte des Projekts. Es soll eine Mensch-Maschine-Interaktion entwickelt werden, die es Mitarbeitenden ermöglicht die KI selbst nachzutrainieren und zu überwachen. Hierdurch ergeben sich neue, zukunftsfähige Arbeitsfelder für die Beschäftigten in Industriewäschereien. Zuerst wurde anhand des Versuchsstandes ein initialer Datensatz aus 9405 Bildern erstellt. Zu jedem der Bilder existieren Informationen über die Farbe, den Typ und zumeist auch über die Kontaminierung, die Schäden und das Material. Anschließend wurden neuronale Netze sowohl zur Extraktion der Region des Wäschestücks als auch zu der Klassifikation der Waschkategorien trainiert, d.h. die Netze lernten zu erkennen, was zum Wäschestück gehört und was nicht und Wäschestücke, Farben und Ver-schmutzungen zu erkennen. Durch einen am aktiven Lernen orientierten Prozess konnten dabei 2091 Segmentierungen automatisch erstellt werden, d.h. ohne menschliche Eingabe der Annotationen. Für die Optimierung dieser neuronalen Netze wurden zusätzlich auch noch vortrainierte Netze nachtrainiert, welches allgemein als Transfer Learning (transferiertes Lernen) bezeichnet wird. Hinsichtlich des Aktiven Lernens wurde der CEAL-Algorithmus erläutert und damit die Sicherheit der Netze bei der Klassifikation bestimmt. Für visuelle Erklärungen wurde Layerwise Relevance Propagation umgesetzt. Wir haben erste Erkenntnisse bzgl. der Sortierung erlangen können und für einige Kate-gorien bereits sehr zufriedenstellende Ergebnisse erlangt. Durch die beiden neuronalen Netze konnten gute Erkennungsgenauigkeiten für alle Kategorien, insbesondere der Farbe, erzielt werden. Hierbei muss jedoch beachtet werden, dass die Datensätze und dementsprechend vor allem die Validierungsdatensätze klein sind. Es ist unbedingt notwendig, den Datensatz zu vergrößern. Dabei sollte höchste Priorität sein, die Aufnahme und Annotation in einer Wäscherei zu verlagern, um möglichst automatisiert und möglichst schnell Daten aufnehmen und annotieren zu können.