Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches
Identifying fungal clones propagated during outbreaks in hospital settings is a problem that increasingly confronts biologists. Current tools based on DNA sequencing or microsatellite analysis require specific manipulations that are difficult to implement in the context of routine diagnosis. Using d...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Microorganisms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-2607/11/4/1071 |
| _version_ | 1797604211071909888 |
|---|---|
| author | Noshine Mohammad Anne-Cécile Normand Cécile Nabet Alexandre Godmer Jean-Yves Brossas Marion Blaize Christine Bonnal Arnaud Fekkar Sébastien Imbert Xavier Tannier Renaud Piarroux |
| author_facet | Noshine Mohammad Anne-Cécile Normand Cécile Nabet Alexandre Godmer Jean-Yves Brossas Marion Blaize Christine Bonnal Arnaud Fekkar Sébastien Imbert Xavier Tannier Renaud Piarroux |
| author_sort | Noshine Mohammad |
| collection | DOAJ |
| description | Identifying fungal clones propagated during outbreaks in hospital settings is a problem that increasingly confronts biologists. Current tools based on DNA sequencing or microsatellite analysis require specific manipulations that are difficult to implement in the context of routine diagnosis. Using deep learning to classify the mass spectra obtained during the routine identification of fungi by MALDI-TOF mass spectrometry could be of interest to differentiate isolates belonging to epidemic clones from others. As part of the management of a nosocomial outbreak due to <i>Candida parapsilosis</i> in two Parisian hospitals, we studied the impact of the preparation of the spectra on the performance of a deep neural network. Our purpose was to differentiate 39 otherwise fluconazole-resistant isolates belonging to a clonal subset from 56 other isolates, most of which were fluconazole-susceptible, collected during the same period and not belonging to the clonal subset. Our study carried out on spectra obtained on four different machines from isolates cultured for 24 or 48 h on three different culture media showed that each of these parameters had a significant impact on the performance of the classifier. In particular, using different culture times between learning and testing steps could lead to a collapse in the accuracy of the predictions. On the other hand, including spectra obtained after 24 and 48 h of growth during the learning step restored the good results. Finally, we showed that the deleterious effect of the device variability used for learning and testing could be largely improved by including a spectra alignment step during preprocessing before submitting them to the neural network. Taken together, these experiments show the great potential of deep learning models to identify spectra of specific clones, providing that crucial parameters are controlled during both culture and preparation steps before submitting spectra to a classifier. |
| first_indexed | 2024-03-11T04:43:03Z |
| format | Article |
| id | doaj.art-f667bfb3bc0e4eb39a4a695dd82b160c |
| institution | Directory Open Access Journal |
| issn | 2076-2607 |
| language | English |
| last_indexed | 2024-03-11T04:43:03Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Microorganisms |
| spelling | doaj.art-f667bfb3bc0e4eb39a4a695dd82b160c2023-11-17T20:34:34ZengMDPI AGMicroorganisms2076-26072023-04-01114107110.3390/microorganisms11041071Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning ApproachesNoshine Mohammad0Anne-Cécile Normand1Cécile Nabet2Alexandre Godmer3Jean-Yves Brossas4Marion Blaize5Christine Bonnal6Arnaud Fekkar7Sébastien Imbert8Xavier Tannier9Renaud Piarroux10Groupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceCIMI-Paris, Centre d’Immunologie et des Maladies Infectieuses, UMR 1135, Sorbonne Université, 75013 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceService de Parasitologie Mycologie, Hôpital Bichat-Claude Bernard, AP-HP, 75018 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceService de Parasitologie Mycologie, Centre Hospitalier Universitaire de Bordeaux, 33075 Bordeaux, FranceSorbonne Université, Inserm, Laboratoire d’Informatique Médicale et d’Ingénierie des Connaissances en e-Santé, LIMICS, 75013 Paris, FranceGroupe Hospitalier Pitié-Salpêtrière, Service de Parasitologie Mycologie, AP-HP, 75013 Paris, FranceIdentifying fungal clones propagated during outbreaks in hospital settings is a problem that increasingly confronts biologists. Current tools based on DNA sequencing or microsatellite analysis require specific manipulations that are difficult to implement in the context of routine diagnosis. Using deep learning to classify the mass spectra obtained during the routine identification of fungi by MALDI-TOF mass spectrometry could be of interest to differentiate isolates belonging to epidemic clones from others. As part of the management of a nosocomial outbreak due to <i>Candida parapsilosis</i> in two Parisian hospitals, we studied the impact of the preparation of the spectra on the performance of a deep neural network. Our purpose was to differentiate 39 otherwise fluconazole-resistant isolates belonging to a clonal subset from 56 other isolates, most of which were fluconazole-susceptible, collected during the same period and not belonging to the clonal subset. Our study carried out on spectra obtained on four different machines from isolates cultured for 24 or 48 h on three different culture media showed that each of these parameters had a significant impact on the performance of the classifier. In particular, using different culture times between learning and testing steps could lead to a collapse in the accuracy of the predictions. On the other hand, including spectra obtained after 24 and 48 h of growth during the learning step restored the good results. Finally, we showed that the deleterious effect of the device variability used for learning and testing could be largely improved by including a spectra alignment step during preprocessing before submitting them to the neural network. Taken together, these experiments show the great potential of deep learning models to identify spectra of specific clones, providing that crucial parameters are controlled during both culture and preparation steps before submitting spectra to a classifier.https://www.mdpi.com/2076-2607/11/4/1071MALDI TOFepidemiology<i>Candida parapsilosis</i>neural networkartificial intelligenceoutbreak |
| spellingShingle | Noshine Mohammad Anne-Cécile Normand Cécile Nabet Alexandre Godmer Jean-Yves Brossas Marion Blaize Christine Bonnal Arnaud Fekkar Sébastien Imbert Xavier Tannier Renaud Piarroux Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches Microorganisms MALDI TOF epidemiology <i>Candida parapsilosis</i> neural network artificial intelligence outbreak |
| title | Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches |
| title_full | Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches |
| title_fullStr | Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches |
| title_full_unstemmed | Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches |
| title_short | Improving the Detection of Epidemic Clones in <i>Candida parapsilosis</i> Outbreaks by Combining MALDI-TOF Mass Spectrometry and Deep Learning Approaches |
| title_sort | improving the detection of epidemic clones in i candida parapsilosis i outbreaks by combining maldi tof mass spectrometry and deep learning approaches |
| topic | MALDI TOF epidemiology <i>Candida parapsilosis</i> neural network artificial intelligence outbreak |
| url | https://www.mdpi.com/2076-2607/11/4/1071 |
| work_keys_str_mv | AT noshinemohammad improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT annececilenormand improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT cecilenabet improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT alexandregodmer improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT jeanyvesbrossas improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT marionblaize improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT christinebonnal improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT arnaudfekkar improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT sebastienimbert improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT xaviertannier improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches AT renaudpiarroux improvingthedetectionofepidemicclonesinicandidaparapsilosisioutbreaksbycombiningmalditofmassspectrometryanddeeplearningapproaches |