A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests
The identification of genomic alterations in tumor tissues, including somatic mutations, deletions, and gene amplifications, produces large amounts of data, which can be correlated with a diversity of therapeutic responses. We aimed to provide a methodological framework to discover pharmacogenomic i...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-06-01
|
| Series: | Genes |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4425/12/6/933 |
| _version_ | 1797529674236035072 |
|---|---|
| author | Salvatore Fasola Giovanna Cilluffo Laura Montalbano Velia Malizia Giuliana Ferrante Stefania La Grutta |
| author_facet | Salvatore Fasola Giovanna Cilluffo Laura Montalbano Velia Malizia Giuliana Ferrante Stefania La Grutta |
| author_sort | Salvatore Fasola |
| collection | DOAJ |
| description | The identification of genomic alterations in tumor tissues, including somatic mutations, deletions, and gene amplifications, produces large amounts of data, which can be correlated with a diversity of therapeutic responses. We aimed to provide a methodological framework to discover pharmacogenomic interactions based on Random Forests. We matched two databases from the Cancer Cell Line Encyclopaedia (CCLE) project, and the Genomics of Drug Sensitivity in Cancer (GDSC) project. For a total of 648 shared cell lines, we considered 48,270 gene alterations from CCLE as input features and the area under the dose-response curve (AUC) for 265 drugs from GDSC as the outcomes. A three-step reduction to 501 alterations was performed, selecting known driver genes and excluding very frequent/infrequent alterations and redundant ones. For each model, we used the concordance correlation coefficient (CCC) for assessing the predictive performance, and permutation importance for assessing the contribution of each alteration. In a reasonable computational time (56 min), we identified 12 compounds whose response was at least fairly sensitive (CCC > 20) to the alteration profiles. Some diversities were found in the sets of influential alterations, providing clues to discover significant drug-gene interactions. The proposed methodological framework can be helpful for mining pharmacogenomic interactions. |
| first_indexed | 2024-03-10T10:17:08Z |
| format | Article |
| id | doaj.art-fd09dc1820454361ba3f2a4bd4aba017 |
| institution | Directory Open Access Journal |
| issn | 2073-4425 |
| language | English |
| last_indexed | 2024-03-10T10:17:08Z |
| publishDate | 2021-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Genes |
| spelling | doaj.art-fd09dc1820454361ba3f2a4bd4aba0172023-11-22T00:45:43ZengMDPI AGGenes2073-44252021-06-0112693310.3390/genes12060933A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random ForestsSalvatore Fasola0Giovanna Cilluffo1Laura Montalbano2Velia Malizia3Giuliana Ferrante4Stefania La Grutta5Institute for Biomedical Research and Innovation, National Research Council, 90146 Palermo, ItalyInstitute for Biomedical Research and Innovation, National Research Council, 90146 Palermo, ItalyInstitute for Biomedical Research and Innovation, National Research Council, 90146 Palermo, ItalyInstitute for Biomedical Research and Innovation, National Research Council, 90146 Palermo, ItalyDepartment of Health Promotion, Mother and Child Care, Internal Medicine and Medical Specialties, University of Palermo, 90127 Palermo, ItalyInstitute for Biomedical Research and Innovation, National Research Council, 90146 Palermo, ItalyThe identification of genomic alterations in tumor tissues, including somatic mutations, deletions, and gene amplifications, produces large amounts of data, which can be correlated with a diversity of therapeutic responses. We aimed to provide a methodological framework to discover pharmacogenomic interactions based on Random Forests. We matched two databases from the Cancer Cell Line Encyclopaedia (CCLE) project, and the Genomics of Drug Sensitivity in Cancer (GDSC) project. For a total of 648 shared cell lines, we considered 48,270 gene alterations from CCLE as input features and the area under the dose-response curve (AUC) for 265 drugs from GDSC as the outcomes. A three-step reduction to 501 alterations was performed, selecting known driver genes and excluding very frequent/infrequent alterations and redundant ones. For each model, we used the concordance correlation coefficient (CCC) for assessing the predictive performance, and permutation importance for assessing the contribution of each alteration. In a reasonable computational time (56 min), we identified 12 compounds whose response was at least fairly sensitive (CCC > 20) to the alteration profiles. Some diversities were found in the sets of influential alterations, providing clues to discover significant drug-gene interactions. The proposed methodological framework can be helpful for mining pharmacogenomic interactions.https://www.mdpi.com/2073-4425/12/6/933cancercell linesdrug responsegenomic alterationspharmacogenomic interactionsRandom Forests |
| spellingShingle | Salvatore Fasola Giovanna Cilluffo Laura Montalbano Velia Malizia Giuliana Ferrante Stefania La Grutta A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests Genes cancer cell lines drug response genomic alterations pharmacogenomic interactions Random Forests |
| title | A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests |
| title_full | A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests |
| title_fullStr | A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests |
| title_full_unstemmed | A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests |
| title_short | A Methodological Framework to Discover Pharmacogenomic Interactions Based on Random Forests |
| title_sort | methodological framework to discover pharmacogenomic interactions based on random forests |
| topic | cancer cell lines drug response genomic alterations pharmacogenomic interactions Random Forests |
| url | https://www.mdpi.com/2073-4425/12/6/933 |
| work_keys_str_mv | AT salvatorefasola amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT giovannacilluffo amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT lauramontalbano amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT veliamalizia amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT giulianaferrante amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT stefanialagrutta amethodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT salvatorefasola methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT giovannacilluffo methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT lauramontalbano methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT veliamalizia methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT giulianaferrante methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests AT stefanialagrutta methodologicalframeworktodiscoverpharmacogenomicinteractionsbasedonrandomforests |