DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis
<p><strong>Motivation</strong> Resistance co-occurrence within first-line anti-tuberculosis (TB) drugs is a common phenomenon. Existing methods based on genetic data analysis of Mycobacterium tuberculosis (MTB) have been able to predict resistance of MTB to individual drugs, but ha...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Published: |
Oxford University Press
2019
|
| _version_ | 1826267923743768576 |
|---|---|
| author | Yang, Y Walker, T Walker, A Wilson, D Peto, T Crook, D Shamout, F Cryptic Consortium Zhu, T Clifton, D |
| author_facet | Yang, Y Walker, T Walker, A Wilson, D Peto, T Crook, D Shamout, F Cryptic Consortium Zhu, T Clifton, D |
| author_sort | Yang, Y |
| collection | OXFORD |
| description | <p><strong>Motivation</strong> Resistance co-occurrence within first-line anti-tuberculosis (TB) drugs is a common phenomenon. Existing methods based on genetic data analysis of Mycobacterium tuberculosis (MTB) have been able to predict resistance of MTB to individual drugs, but have not considered the resistance co-occurrence and cannot capture latent structure of genomic data that corresponds to lineages.</p> <p><strong>Results</strong> We used a large cohort of TB patients from 16 countries across six continents where whole-genome sequences for each isolate and associated phenotype to anti-TB drugs were obtained using drug susceptibility testing recommended by the World Health Organization. We then proposed an end-to-end multi-task model with deep denoising auto-encoder (DeepAMR) for multiple drug classification and developed DeepAMR_cluster, a clustering variant based on DeepAMR, for learning clusters in latent space of the data. The results showed that DeepAMR outperformed baseline model and four machine learning models with mean AUROC from 94.4% to 98.7% for predicting resistance to four first-line drugs [i.e. isoniazid (INH), ethambutol (EMB), rifampicin (RIF), pyrazinamide (PZA)], multi-drug resistant TB (MDR-TB) and pan-susceptible TB (PANS-TB: MTB that is susceptible to all four first-line anti-TB drugs). In the case of INH, EMB, PZA and MDR-TB, DeepAMR achieved its best mean sensitivity of 94.3%, 91.5%, 87.3% and 96.3%, respectively. While in the case of RIF and PANS-TB, it generated 94.2% and 92.2% sensitivity, which were lower than baseline model by 0.7% and 1.9%, respectively. t-SNE visualization shows that DeepAMR_cluster captures lineage-related clusters in the latent space.</p> <p><strong>Availability</strong> The details of source code are provided at http://www.robots.ox.ac.uk/∼davidc/code.php.</p> |
| first_indexed | 2024-03-06T21:01:43Z |
| format | Journal article |
| id | oxford-uuid:3b1a36ea-ee7f-4d41-a862-bb8ebb7fac4a |
| institution | University of Oxford |
| last_indexed | 2024-03-06T21:01:43Z |
| publishDate | 2019 |
| publisher | Oxford University Press |
| record_format | dspace |
| spelling | oxford-uuid:3b1a36ea-ee7f-4d41-a862-bb8ebb7fac4a2022-03-26T14:05:33ZDeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosisJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:3b1a36ea-ee7f-4d41-a862-bb8ebb7fac4aSymplectic Elements at OxfordOxford University Press2019Yang, YWalker, TWalker, AWilson, DPeto, TCrook, DShamout, FCryptic ConsortiumZhu, TClifton, D<p><strong>Motivation</strong> Resistance co-occurrence within first-line anti-tuberculosis (TB) drugs is a common phenomenon. Existing methods based on genetic data analysis of Mycobacterium tuberculosis (MTB) have been able to predict resistance of MTB to individual drugs, but have not considered the resistance co-occurrence and cannot capture latent structure of genomic data that corresponds to lineages.</p> <p><strong>Results</strong> We used a large cohort of TB patients from 16 countries across six continents where whole-genome sequences for each isolate and associated phenotype to anti-TB drugs were obtained using drug susceptibility testing recommended by the World Health Organization. We then proposed an end-to-end multi-task model with deep denoising auto-encoder (DeepAMR) for multiple drug classification and developed DeepAMR_cluster, a clustering variant based on DeepAMR, for learning clusters in latent space of the data. The results showed that DeepAMR outperformed baseline model and four machine learning models with mean AUROC from 94.4% to 98.7% for predicting resistance to four first-line drugs [i.e. isoniazid (INH), ethambutol (EMB), rifampicin (RIF), pyrazinamide (PZA)], multi-drug resistant TB (MDR-TB) and pan-susceptible TB (PANS-TB: MTB that is susceptible to all four first-line anti-TB drugs). In the case of INH, EMB, PZA and MDR-TB, DeepAMR achieved its best mean sensitivity of 94.3%, 91.5%, 87.3% and 96.3%, respectively. While in the case of RIF and PANS-TB, it generated 94.2% and 92.2% sensitivity, which were lower than baseline model by 0.7% and 1.9%, respectively. t-SNE visualization shows that DeepAMR_cluster captures lineage-related clusters in the latent space.</p> <p><strong>Availability</strong> The details of source code are provided at http://www.robots.ox.ac.uk/∼davidc/code.php.</p> |
| spellingShingle | Yang, Y Walker, T Walker, A Wilson, D Peto, T Crook, D Shamout, F Cryptic Consortium Zhu, T Clifton, D DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title | DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title_full | DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title_fullStr | DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title_full_unstemmed | DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title_short | DeepAMR for predicting co-occurrent resistance of Mycobacterium tuberculosis |
| title_sort | deepamr for predicting co occurrent resistance of mycobacterium tuberculosis |
| work_keys_str_mv | AT yangy deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT walkert deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT walkera deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT wilsond deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT petot deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT crookd deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT shamoutf deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT crypticconsortium deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT zhut deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis AT cliftond deepamrforpredictingcooccurrentresistanceofmycobacteriumtuberculosis |