Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A Deep Learning Pipeline for Genome-Wide Imaging Screen Uncovering Cell Death Regulators

Kazemi Rashed, Salma LU ; Esbo, Klara ; Miari, Mariam LU ; Arvidsson, Malou and Aits, Sonja LU orcid (2025) 2024 International Conference on Bioinformatics and Biomedicine (BIBM) p.7123-7123
Abstract
Abstract—In numerous image-based studies, such as cell studies,
a plethora of information is embedded in cell objects, which
must first be accurately extracted before it can be utilized for
further downstream analysis and linked to biological processes
explored in the original experiment. One such technique for
this purpose is instance segmentation. In our study, which
cell death and associated pathways and regulatory networks
are investigated through a genome-wide microscopic screen,
the dataset is extensive and challenging to manually curate
or annotate. Therefore, we opted to train distinct U-shape
nuclei segmentation models using only a small but meticulously
annotated subset of the... (More)
Abstract—In numerous image-based studies, such as cell studies,
a plethora of information is embedded in cell objects, which
must first be accurately extracted before it can be utilized for
further downstream analysis and linked to biological processes
explored in the original experiment. One such technique for
this purpose is instance segmentation. In our study, which
cell death and associated pathways and regulatory networks
are investigated through a genome-wide microscopic screen,
the dataset is extensive and challenging to manually curate
or annotate. Therefore, we opted to train distinct U-shape
nuclei segmentation models using only a small but meticulously
annotated subset of the high-content screen data to ensure high
accuracy where the superior U-Net model achieved an F1-score
of 89%. Subsequently, we have deployed the best U-Net model
to segment whole objects in the nuclei channel of the original
genome-wide dataset, comprising nearly 5 million image frames
across 213 plates. The extracted features from the nuclei channel
served as biological metrics to explore cell death. The results
identified several well-known genes involved in established cell
death pathways, such as apoptosis. Additionally, other genes were
discovered that merit further investigation for their effects on
altering the visual features of cell nuclei.
Index Terms—Cell death pathways, Image Segmentation,
Genome-wide study, U-Net. (Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Cell death, U-Net, Instance segmentation, High-content imaging, artificial intelligence, deep learning, computer vision
host publication
2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
pages
7123 - 7123
publisher
IEEE
conference name
2024 International Conference on Bioinformatics and Biomedicine (BIBM)
conference location
Lisbon, Portugal
conference dates
2024-12-03 - 2024-12-06
external identifiers
  • scopus:85217276976
ISBN
979-8-3503-8622-6
979-8-3503-8623-3
DOI
10.1109/BIBM62325.2024.10822022
project
Lund AI Research - Development of AI Technologies
Lysosomes in cell death - from molecular mechanisms to new treatment strategies
Revealing drivers of cell death disruption across species and theri links to biodiversity loss and human disease
Analysing large-scale microscopy datasets with AI-based approaches
Understanding and therapeutic targeting of lysosome-dependent cell death
Environmental toxins as cause of lysosomal damage-induced cell death in animals and humans
Using artificial intelligence and advanced computational tools to identify biological pathways, disease mechanisms and therapeutic opportunities
language
English
LU publication?
yes
id
5b90c5ea-6be3-49aa-b447-a55cdeb33d12
date added to LUP
2024-12-13 12:32:12
date last changed
2026-01-16 03:28:35
@inproceedings{5b90c5ea-6be3-49aa-b447-a55cdeb33d12,
  abstract     = {{Abstract—In numerous image-based studies, such as cell studies,<br/>a plethora of information is embedded in cell objects, which<br/>must first be accurately extracted before it can be utilized for<br/>further downstream analysis and linked to biological processes<br/>explored in the original experiment. One such technique for<br/>this purpose is instance segmentation. In our study, which<br/>cell death and associated pathways and regulatory networks<br/>are investigated through a genome-wide microscopic screen,<br/>the dataset is extensive and challenging to manually curate<br/>or annotate. Therefore, we opted to train distinct U-shape<br/>nuclei segmentation models using only a small but meticulously<br/>annotated subset of the high-content screen data to ensure high<br/>accuracy where the superior U-Net model achieved an F1-score<br/>of 89%. Subsequently, we have deployed the best U-Net model<br/>to segment whole objects in the nuclei channel of the original<br/>genome-wide dataset, comprising nearly 5 million image frames<br/>across 213 plates. The extracted features from the nuclei channel<br/>served as biological metrics to explore cell death. The results<br/>identified several well-known genes involved in established cell<br/>death pathways, such as apoptosis. Additionally, other genes were<br/>discovered that merit further investigation for their effects on<br/>altering the visual features of cell nuclei.<br/>Index Terms—Cell death pathways, Image Segmentation,<br/>Genome-wide study, U-Net.}},
  author       = {{Kazemi Rashed, Salma and Esbo, Klara and Miari, Mariam and Arvidsson, Malou and Aits, Sonja}},
  booktitle    = {{2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)}},
  isbn         = {{979-8-3503-8622-6}},
  keywords     = {{Cell death; U-Net; Instance segmentation; High-content imaging; artificial intelligence; deep learning; computer vision}},
  language     = {{eng}},
  month        = {{01}},
  pages        = {{7123--7123}},
  publisher    = {{IEEE}},
  title        = {{A Deep Learning Pipeline for Genome-Wide Imaging Screen Uncovering Cell Death Regulators}},
  url          = {{http://dx.doi.org/10.1109/BIBM62325.2024.10822022}},
  doi          = {{10.1109/BIBM62325.2024.10822022}},
  year         = {{2025}},
}