Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach
Aquaculture located in urban river estuaries, where other anthropogenic activities may occur, has an impact on and may be affected by the environment where they are inserted, namely by the exchange of antimicrobial resistance genes. The latter may ultimately, through the food chain, represent a sour...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Antibiotics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-6382/13/1/107 |
| _version_ | 1797344771097755648 |
|---|---|
| author | Helena Sofia Salgueiro Ana Cristina Ferreira Ana Sofia Ribeiro Duarte Ana Botelho |
| author_facet | Helena Sofia Salgueiro Ana Cristina Ferreira Ana Sofia Ribeiro Duarte Ana Botelho |
| author_sort | Helena Sofia Salgueiro |
| collection | DOAJ |
| description | Aquaculture located in urban river estuaries, where other anthropogenic activities may occur, has an impact on and may be affected by the environment where they are inserted, namely by the exchange of antimicrobial resistance genes. The latter may ultimately, through the food chain, represent a source of resistance genes to the human resistome. In an exploratory study of the presence of resistance genes in aquaculture sediments located in urban river estuaries, two machine learning models were applied to predict the source of 34 resistome observations in the aquaculture sediments of oysters and gilt-head sea bream, located in the estuaries of the Sado and Lima Rivers and in the Aveiro Lagoon, as well as in the sediments of the Tejo River estuary, where Japanese clams and mussels are collected. The first model included all 34 resistomes, amounting to 53 different antimicrobial resistance genes used as source predictors. The most important antimicrobial genes for source attribution were tetracycline resistance genes <i>tet</i>(<i>51</i>) and <i>tet</i>(<i>L</i>); aminoglycoside resistance gene <i>aadA6</i>; beta-lactam resistance gene <i>blaBRO-2</i>; and amphenicol resistance gene <i>cmx_1</i>. The second model included only oyster sediment resistomes, amounting to 30 antimicrobial resistance genes as predictors. The most important antimicrobial genes for source attribution were the aminoglycoside resistance gene <i>aadA6</i>, followed by the tetracycline genes <i>tet</i>(<i>L</i>) and <i>tet</i>(<i>33</i>). This exploratory study provides the first information about antimicrobial resistance genes in intensive and semi-intensive aquaculture in Portugal, helping to recognize the importance of environmental control to maintain the integrity and the sustainability of aquaculture farms. |
| first_indexed | 2024-03-08T11:07:39Z |
| format | Article |
| id | doaj.art-b8b78bafd2464ab29cd622772501f408 |
| institution | Directory Open Access Journal |
| issn | 2079-6382 |
| language | English |
| last_indexed | 2024-03-08T11:07:39Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Antibiotics |
| spelling | doaj.art-b8b78bafd2464ab29cd622772501f4082024-01-26T14:39:05ZengMDPI AGAntibiotics2079-63822024-01-0113110710.3390/antibiotics13010107Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning ApproachHelena Sofia Salgueiro0Ana Cristina Ferreira1Ana Sofia Ribeiro Duarte2Ana Botelho3Faculty of Veterinary Medicine, University of Lisbon, 1300-477 Lisbon, PortugalNational Institute for Agrarian and Veterinary Research (INIAV IP), Av. da República, Quinta do Marquês, 2780-157 Oeiras, PortugalNational Food Institute, Technical University of Denmark, Kemitorvet 204, 2800 Kongens Lyngby, DenmarkNational Institute for Agrarian and Veterinary Research (INIAV IP), Av. da República, Quinta do Marquês, 2780-157 Oeiras, PortugalAquaculture located in urban river estuaries, where other anthropogenic activities may occur, has an impact on and may be affected by the environment where they are inserted, namely by the exchange of antimicrobial resistance genes. The latter may ultimately, through the food chain, represent a source of resistance genes to the human resistome. In an exploratory study of the presence of resistance genes in aquaculture sediments located in urban river estuaries, two machine learning models were applied to predict the source of 34 resistome observations in the aquaculture sediments of oysters and gilt-head sea bream, located in the estuaries of the Sado and Lima Rivers and in the Aveiro Lagoon, as well as in the sediments of the Tejo River estuary, where Japanese clams and mussels are collected. The first model included all 34 resistomes, amounting to 53 different antimicrobial resistance genes used as source predictors. The most important antimicrobial genes for source attribution were tetracycline resistance genes <i>tet</i>(<i>51</i>) and <i>tet</i>(<i>L</i>); aminoglycoside resistance gene <i>aadA6</i>; beta-lactam resistance gene <i>blaBRO-2</i>; and amphenicol resistance gene <i>cmx_1</i>. The second model included only oyster sediment resistomes, amounting to 30 antimicrobial resistance genes as predictors. The most important antimicrobial genes for source attribution were the aminoglycoside resistance gene <i>aadA6</i>, followed by the tetracycline genes <i>tet</i>(<i>L</i>) and <i>tet</i>(<i>33</i>). This exploratory study provides the first information about antimicrobial resistance genes in intensive and semi-intensive aquaculture in Portugal, helping to recognize the importance of environmental control to maintain the integrity and the sustainability of aquaculture farms.https://www.mdpi.com/2079-6382/13/1/107aquaculture sedimentsantimicrobial resistanceresistomessource attributionmachine learning |
| spellingShingle | Helena Sofia Salgueiro Ana Cristina Ferreira Ana Sofia Ribeiro Duarte Ana Botelho Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach Antibiotics aquaculture sediments antimicrobial resistance resistomes source attribution machine learning |
| title | Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach |
| title_full | Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach |
| title_fullStr | Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach |
| title_full_unstemmed | Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach |
| title_short | Source Attribution of Antibiotic Resistance Genes in Estuarine Aquaculture: A Machine Learning Approach |
| title_sort | source attribution of antibiotic resistance genes in estuarine aquaculture a machine learning approach |
| topic | aquaculture sediments antimicrobial resistance resistomes source attribution machine learning |
| url | https://www.mdpi.com/2079-6382/13/1/107 |
| work_keys_str_mv | AT helenasofiasalgueiro sourceattributionofantibioticresistancegenesinestuarineaquacultureamachinelearningapproach AT anacristinaferreira sourceattributionofantibioticresistancegenesinestuarineaquacultureamachinelearningapproach AT anasofiaribeiroduarte sourceattributionofantibioticresistancegenesinestuarineaquacultureamachinelearningapproach AT anabotelho sourceattributionofantibioticresistancegenesinestuarineaquacultureamachinelearningapproach |