Not getting in too deep: A practical deep learning approach to routine crystallisation image classification
Using a relatively small training set of ~16 thousand images from macromolecular crystallisation experiments, we compare classification results obtained with four of the most widely-used convolutional deep-learning network architectures that can be implemented without the need for extensive computat...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9997964/?tool=EBI |
| _version_ | 1811154300629417984 |
|---|---|
| author | Jamie Milne Chen Qian David Hargreaves Yinhai Wang Julie Wilson |
| author_facet | Jamie Milne Chen Qian David Hargreaves Yinhai Wang Julie Wilson |
| author_sort | Jamie Milne |
| collection | DOAJ |
| description | Using a relatively small training set of ~16 thousand images from macromolecular crystallisation experiments, we compare classification results obtained with four of the most widely-used convolutional deep-learning network architectures that can be implemented without the need for extensive computational resources. We show that the classifiers have different strengths that can be combined to provide an ensemble classifier achieving a classification accuracy comparable to that obtained by a large consortium initiative. We use eight classes to effectively rank the experimental outcomes, thereby providing detailed information that can be used with routine crystallography experiments to automatically identify crystal formation for drug discovery and pave the way for further exploration of the relationship between crystal formation and crystallisation conditions. |
| first_indexed | 2024-04-10T04:13:26Z |
| format | Article |
| id | doaj.art-0e4e347d0660415786e59f1c92f588d0 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-10T04:13:26Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-0e4e347d0660415786e59f1c92f588d02023-03-12T05:32:20ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01183Not getting in too deep: A practical deep learning approach to routine crystallisation image classificationJamie MilneChen QianDavid HargreavesYinhai WangJulie WilsonUsing a relatively small training set of ~16 thousand images from macromolecular crystallisation experiments, we compare classification results obtained with four of the most widely-used convolutional deep-learning network architectures that can be implemented without the need for extensive computational resources. We show that the classifiers have different strengths that can be combined to provide an ensemble classifier achieving a classification accuracy comparable to that obtained by a large consortium initiative. We use eight classes to effectively rank the experimental outcomes, thereby providing detailed information that can be used with routine crystallography experiments to automatically identify crystal formation for drug discovery and pave the way for further exploration of the relationship between crystal formation and crystallisation conditions.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9997964/?tool=EBI |
| spellingShingle | Jamie Milne Chen Qian David Hargreaves Yinhai Wang Julie Wilson Not getting in too deep: A practical deep learning approach to routine crystallisation image classification PLoS ONE |
| title | Not getting in too deep: A practical deep learning approach to routine crystallisation image classification |
| title_full | Not getting in too deep: A practical deep learning approach to routine crystallisation image classification |
| title_fullStr | Not getting in too deep: A practical deep learning approach to routine crystallisation image classification |
| title_full_unstemmed | Not getting in too deep: A practical deep learning approach to routine crystallisation image classification |
| title_short | Not getting in too deep: A practical deep learning approach to routine crystallisation image classification |
| title_sort | not getting in too deep a practical deep learning approach to routine crystallisation image classification |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9997964/?tool=EBI |
| work_keys_str_mv | AT jamiemilne notgettingintoodeepapracticaldeeplearningapproachtoroutinecrystallisationimageclassification AT chenqian notgettingintoodeepapracticaldeeplearningapproachtoroutinecrystallisationimageclassification AT davidhargreaves notgettingintoodeepapracticaldeeplearningapproachtoroutinecrystallisationimageclassification AT yinhaiwang notgettingintoodeepapracticaldeeplearningapproachtoroutinecrystallisationimageclassification AT juliewilson notgettingintoodeepapracticaldeeplearningapproachtoroutinecrystallisationimageclassification |