Cell-specific imputation of drug connectivity mapping with incomplete data.
Drug repositioning allows expedited discovery of new applications for existing compounds, but re-screening vast compound libraries is often prohibitively expensive. "Connectivity mapping" is a process that links drugs to diseases by identifying compounds whose impact on expression in a col...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0278289 |
| _version_ | 1797852152940462080 |
|---|---|
| author | Diana Sapashnik Rebecca Newman Christopher Michael Pietras Di Zhou Kapil Devkota Fangfang Qu Lior Kofman Sean Boudreau Inbar Fried Donna K Slonim |
| author_facet | Diana Sapashnik Rebecca Newman Christopher Michael Pietras Di Zhou Kapil Devkota Fangfang Qu Lior Kofman Sean Boudreau Inbar Fried Donna K Slonim |
| author_sort | Diana Sapashnik |
| collection | DOAJ |
| description | Drug repositioning allows expedited discovery of new applications for existing compounds, but re-screening vast compound libraries is often prohibitively expensive. "Connectivity mapping" is a process that links drugs to diseases by identifying compounds whose impact on expression in a collection of cells reverses the disease's impact on expression in disease-relevant tissues. The LINCS project has expanded the universe of compounds and cells for which data are available, but even with this effort, many clinically useful combinations are missing. To evaluate the possibility of repurposing drugs despite missing data, we compared collaborative filtering using either neighborhood-based or SVD imputation methods to two naive approaches via cross-validation. Methods were evaluated for their ability to predict drug connectivity despite missing data. Predictions improved when cell type was taken into account. Neighborhood collaborative filtering was the most successful method, with the best improvements in non-immortalized primary cells. We also explored which classes of compounds are most and least reliant on cell type for accurate imputation. We conclude that even for cells in which drug responses have not been fully characterized, it is possible to identify unassayed drugs that reverse in those cells the expression signatures observed in disease. |
| first_indexed | 2024-04-09T19:29:27Z |
| format | Article |
| id | doaj.art-fcb9b705d9e24db8a3db073122fb041f |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-09T19:29:27Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-fcb9b705d9e24db8a3db073122fb041f2023-04-05T05:31:28ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01182e027828910.1371/journal.pone.0278289Cell-specific imputation of drug connectivity mapping with incomplete data.Diana SapashnikRebecca NewmanChristopher Michael PietrasDi ZhouKapil DevkotaFangfang QuLior KofmanSean BoudreauInbar FriedDonna K SlonimDrug repositioning allows expedited discovery of new applications for existing compounds, but re-screening vast compound libraries is often prohibitively expensive. "Connectivity mapping" is a process that links drugs to diseases by identifying compounds whose impact on expression in a collection of cells reverses the disease's impact on expression in disease-relevant tissues. The LINCS project has expanded the universe of compounds and cells for which data are available, but even with this effort, many clinically useful combinations are missing. To evaluate the possibility of repurposing drugs despite missing data, we compared collaborative filtering using either neighborhood-based or SVD imputation methods to two naive approaches via cross-validation. Methods were evaluated for their ability to predict drug connectivity despite missing data. Predictions improved when cell type was taken into account. Neighborhood collaborative filtering was the most successful method, with the best improvements in non-immortalized primary cells. We also explored which classes of compounds are most and least reliant on cell type for accurate imputation. We conclude that even for cells in which drug responses have not been fully characterized, it is possible to identify unassayed drugs that reverse in those cells the expression signatures observed in disease.https://doi.org/10.1371/journal.pone.0278289 |
| spellingShingle | Diana Sapashnik Rebecca Newman Christopher Michael Pietras Di Zhou Kapil Devkota Fangfang Qu Lior Kofman Sean Boudreau Inbar Fried Donna K Slonim Cell-specific imputation of drug connectivity mapping with incomplete data. PLoS ONE |
| title | Cell-specific imputation of drug connectivity mapping with incomplete data. |
| title_full | Cell-specific imputation of drug connectivity mapping with incomplete data. |
| title_fullStr | Cell-specific imputation of drug connectivity mapping with incomplete data. |
| title_full_unstemmed | Cell-specific imputation of drug connectivity mapping with incomplete data. |
| title_short | Cell-specific imputation of drug connectivity mapping with incomplete data. |
| title_sort | cell specific imputation of drug connectivity mapping with incomplete data |
| url | https://doi.org/10.1371/journal.pone.0278289 |
| work_keys_str_mv | AT dianasapashnik cellspecificimputationofdrugconnectivitymappingwithincompletedata AT rebeccanewman cellspecificimputationofdrugconnectivitymappingwithincompletedata AT christophermichaelpietras cellspecificimputationofdrugconnectivitymappingwithincompletedata AT dizhou cellspecificimputationofdrugconnectivitymappingwithincompletedata AT kapildevkota cellspecificimputationofdrugconnectivitymappingwithincompletedata AT fangfangqu cellspecificimputationofdrugconnectivitymappingwithincompletedata AT liorkofman cellspecificimputationofdrugconnectivitymappingwithincompletedata AT seanboudreau cellspecificimputationofdrugconnectivitymappingwithincompletedata AT inbarfried cellspecificimputationofdrugconnectivitymappingwithincompletedata AT donnakslonim cellspecificimputationofdrugconnectivitymappingwithincompletedata |