Automatic classification of fungal-fungal interactions using deep leaning models

Fungi provide valuable solutions for diverse biotechnological applications, such as enzymes in the food industry, bioactive metabolites for healthcare, and biocontrol organisms in agriculture. Current workflows for identifying new biocontrol fungi often rely on subjective visual observations of stra...

Full description

Saved in:
Bibliographic Details
Main Authors: Marjan Mansourvar, Jonathan Funk, Søren Dalsgård Petersen, Sajad Tavakoli, Jakob Blæsbjerg Hoof, David Llorente Corcoles, Sabrina M. Pittroff, Lars Jelsbak, Niels Bjerg Jensen, Ling Ding, Rasmus John Normand Frandsen
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Computational and Structural Biotechnology Journal
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2001037024004008
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846157833456844800
author Marjan Mansourvar
Jonathan Funk
Søren Dalsgård Petersen
Sajad Tavakoli
Jakob Blæsbjerg Hoof
David Llorente Corcoles
Sabrina M. Pittroff
Lars Jelsbak
Niels Bjerg Jensen
Ling Ding
Rasmus John Normand Frandsen
author_facet Marjan Mansourvar
Jonathan Funk
Søren Dalsgård Petersen
Sajad Tavakoli
Jakob Blæsbjerg Hoof
David Llorente Corcoles
Sabrina M. Pittroff
Lars Jelsbak
Niels Bjerg Jensen
Ling Ding
Rasmus John Normand Frandsen
author_sort Marjan Mansourvar
collection DOAJ
description Fungi provide valuable solutions for diverse biotechnological applications, such as enzymes in the food industry, bioactive metabolites for healthcare, and biocontrol organisms in agriculture. Current workflows for identifying new biocontrol fungi often rely on subjective visual observations of strains’ performance in microbe-microbe interaction studies, making the process time-consuming and difficult to reproduce. To overcome these challenges, we developed an AI-automated image classification approach using machine learning algorithm based on deep neural network. Our method focuses on analyzing standardized images of 96-well microtiter plates with solid medium for fungal-fungal challenge experiments. We used our model to categorize the outcome of interactions between the plant pathogen Fusarium graminearum and individual isolates from a collection of 38,400 fungal strains. The authors trained multiple deep learning architectures and evaluated their performance. The results strongly support our approach, achieving a peak accuracy of 95.0 % with the DenseNet121 model and a maximum macro-averaged F1-Score of 93.1 across five folds. To the best of our knowledge, this paper introduces the first automated method for classifying fungal–fungal interactions using deep learning, which can easily be adapted for other fungal species.
format Article
id doaj-art-ac965e91f07b4b6f899f54a9468bc2b4
institution Kabale University
issn 2001-0370
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Computational and Structural Biotechnology Journal
spelling doaj-art-ac965e91f07b4b6f899f54a9468bc2b42024-11-25T04:40:59ZengElsevierComputational and Structural Biotechnology Journal2001-03702024-12-012342224231Automatic classification of fungal-fungal interactions using deep leaning modelsMarjan Mansourvar0Jonathan Funk1Søren Dalsgård Petersen2Sajad Tavakoli3Jakob Blæsbjerg Hoof4David Llorente Corcoles5Sabrina M. Pittroff6Lars Jelsbak7Niels Bjerg Jensen8Ling Ding9Rasmus John Normand Frandsen10Corresponding author.; Department of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkDepartment of Biotechnology and Biomedicine (DTU Bioengineering) Technical University of Denmark, Søltofts Plads, Kgs. Lyngby 2800, DenmarkFungi provide valuable solutions for diverse biotechnological applications, such as enzymes in the food industry, bioactive metabolites for healthcare, and biocontrol organisms in agriculture. Current workflows for identifying new biocontrol fungi often rely on subjective visual observations of strains’ performance in microbe-microbe interaction studies, making the process time-consuming and difficult to reproduce. To overcome these challenges, we developed an AI-automated image classification approach using machine learning algorithm based on deep neural network. Our method focuses on analyzing standardized images of 96-well microtiter plates with solid medium for fungal-fungal challenge experiments. We used our model to categorize the outcome of interactions between the plant pathogen Fusarium graminearum and individual isolates from a collection of 38,400 fungal strains. The authors trained multiple deep learning architectures and evaluated their performance. The results strongly support our approach, achieving a peak accuracy of 95.0 % with the DenseNet121 model and a maximum macro-averaged F1-Score of 93.1 across five folds. To the best of our knowledge, this paper introduces the first automated method for classifying fungal–fungal interactions using deep learning, which can easily be adapted for other fungal species.http://www.sciencedirect.com/science/article/pii/S2001037024004008Fungal growthDeep learningMachine learningComputer visionAutomationBiocontrol
spellingShingle Marjan Mansourvar
Jonathan Funk
Søren Dalsgård Petersen
Sajad Tavakoli
Jakob Blæsbjerg Hoof
David Llorente Corcoles
Sabrina M. Pittroff
Lars Jelsbak
Niels Bjerg Jensen
Ling Ding
Rasmus John Normand Frandsen
Automatic classification of fungal-fungal interactions using deep leaning models
Computational and Structural Biotechnology Journal
Fungal growth
Deep learning
Machine learning
Computer vision
Automation
Biocontrol
title Automatic classification of fungal-fungal interactions using deep leaning models
title_full Automatic classification of fungal-fungal interactions using deep leaning models
title_fullStr Automatic classification of fungal-fungal interactions using deep leaning models
title_full_unstemmed Automatic classification of fungal-fungal interactions using deep leaning models
title_short Automatic classification of fungal-fungal interactions using deep leaning models
title_sort automatic classification of fungal fungal interactions using deep leaning models
topic Fungal growth
Deep learning
Machine learning
Computer vision
Automation
Biocontrol
url http://www.sciencedirect.com/science/article/pii/S2001037024004008
work_keys_str_mv AT marjanmansourvar automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT jonathanfunk automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT sørendalsgardpetersen automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT sajadtavakoli automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT jakobblæsbjerghoof automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT davidllorentecorcoles automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT sabrinampittroff automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT larsjelsbak automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT nielsbjergjensen automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT lingding automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels
AT rasmusjohnnormandfrandsen automaticclassificationoffungalfungalinteractionsusingdeepleaningmodels