Capturing cell heterogeneity in representations of cell populations for image-based profiling using contrastive learning

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Nov. 16, 2023

Image-based cell profiling is a powerful tool that compares perturbed populations by measuring thousands of single-cell features and summarizing them into profiles. Typically sample represented averaging across cells, but this fails to capture the heterogeneity within populations. We introduce CytoSummaryNet: Deep Sets-based approach improves mechanism action prediction 30-68% in mean average precision compared on public dataset. CytoSummaryNet uses self-supervised contrastive learning multiple-instance framework, providing an easier-to-apply method for aggregating feature data than previously published strategies. Interpretability analysis suggests model achieves improvement downweighting small mitotic cells or those with debris prioritizing large uncrowded cells. The requires only perturbation labels training, which are readily available all datasets. offers straightforward post-processing step profiles can significantly boost retrieval performance image-based

Language: Английский

---

Unbiased identification of cell identity in dense mixed neural cultures

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 8, 2024

Abstract Induced pluripotent stem cell (iPSC) technology is revolutionizing biology. However, the variability between individual iPSC lines and lack of efficient to comprehensively characterize iPSC-derived types hinder its adoption in routine preclinical screening settings. To facilitate validation culture composition, we have implemented an imaging assay based on painting convolutional neural networks recognize dense mixed cultures with high fidelity. We benchmarked our approach using pure neuroblastoma astrocytoma attained a classification accuracy above 96%. Through iterative data erosion found that inputs containing nuclear region interest close environment, allow achieving equally as whole for semi-confluent preserved prediction even very cultures. then applied this regionally restricted profiling evaluate differentiation status cultures, by determining ratio postmitotic neurons progenitors. cell-based significantly outperformed which time was used criterion (96% vs. 86%, resp.). In neuronal microglia could be unequivocally discriminated from neurons, regardless their reactivity state. A tiered strategy, allowed discriminating microglial states well, albeit lower accuracy. Thus, morphological single provides means quantify composition complex holds promise use quality control models.

Language: Английский

---

Systematic data analysis pipeline for quantitative morphological cell phenotyping

Computational and Structural Biotechnology Journal, Journal Year: 2024, Volume and Issue: 23, P. 2949 - 2962

Published: July 14, 2024

Quantitative morphological phenotyping (QMP) is an image-based method used to capture features at both the cellular and population level. Its interdisciplinary nature, spanning from data collection result analysis interpretation, can lead uncertainties, particularly among those new this actively growing field. High analytical specificity for a typical QMP achieved through sophisticated approaches that leverage subtle changes. Here, we outline systematic workflow refine methodology. For practical review, describe main steps of QMP; in each step, discuss available methods, their applications, advantages, disadvantages, along with R functions packages easy implementation. This review does not cover theoretical backgrounds, but provides several references interested researchers. It aims broaden horizons future phenome studies demonstrate how exploit years endeavors achieve more less.

Language: Английский

---

Unbiased identification of cell identity in dense mixed neural cultures

Published: Nov. 5, 2024

Induced pluripotent stem cell (iPSC) technology is revolutionizing biology. However, the variability between individual iPSC lines and lack of efficient to comprehensively characterize iPSC-derived types hinder its adoption in routine preclinical screening settings. To facilitate validation culture composition, we have implemented an imaging assay based on painting convolutional neural networks recognize dense mixed cultures with high fidelity. We benchmarked our approach using pure neuroblastoma astrocytoma attained a classification accuracy above 96%. Through iterative data erosion found that inputs containing nuclear region interest close environment, allow achieving equally as whole for semi-confluent preserved prediction even very cultures. then applied this regionally restricted profiling evaluate differentiation status cultures, by determining ratio postmitotic neurons progenitors. cell-based significantly outperformed which time was used criterion (96% vs. 86%, resp.). In neuronal microglia could be unequivocally discriminated from neurons, regardless their reactivity state. A tiered strategy, allowed discriminating microglial states well, albeit lower accuracy. Thus, morphological single provides means quantify composition complex holds promise use quality control models.

Language: Английский

---

A Highly-Efficient, Scalable Pipeline for Fixed Feature Extraction from Large-Scale High-Content Imaging Screens

iScience, Journal Year: 2024, Volume and Issue: 27(12), P. 111434 - 111434

Published: Nov. 21, 2024

Language: Английский

---