Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma
While deep learning (DL) is used in patients’ outcome predictions, the insufficiency of patient samples limits the accuracy. In this study, we investigated how transfer learning (TL) alleviates the small sample size problem. A 2-step TL framework was constructed for a difficult task: predicting the...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SAGE Publishing
2025-01-01
|
| Series: | Bioinformatics and Biology Insights |
| Online Access: | https://doi.org/10.1177/11779322241301507 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841560384890208256 |
|---|---|
| author | Jie Ju Ioannis Ntafoulis Michelle Klein Marcel JT Reinders Martine Lamfers Andrew P Stubbs Yunlei Li |
| author_facet | Jie Ju Ioannis Ntafoulis Michelle Klein Marcel JT Reinders Martine Lamfers Andrew P Stubbs Yunlei Li |
| author_sort | Jie Ju |
| collection | DOAJ |
| description | While deep learning (DL) is used in patients’ outcome predictions, the insufficiency of patient samples limits the accuracy. In this study, we investigated how transfer learning (TL) alleviates the small sample size problem. A 2-step TL framework was constructed for a difficult task: predicting the response of the drug temozolomide (TMZ) in glioblastoma (GBM) cell cultures. The GBM is aggressive, and most patients do not benefit from the only approved chemotherapeutic agent TMZ. O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status is the only biomarker for TMZ responsiveness but has shown limited predictive power. The 2-step TL framework was built on 3 datasets: (1) the subset of the Genomics of Drug Sensitivity in Cancer (GDSC) dataset, including miscellaneous cell cultures treated by TMZ, cyclophosphamide, bortezomib, and oxaliplatin, as the source dataset; (2) the Human Glioblastoma Cell Culture (HGCC) dataset, for fine-tuning; and (3) a small target dataset GSE232173, for validation. The latter two included specifically TMZ-treated GBM cell cultures. The DL models were pretrained on the cell cultures treated by each of the 4 drugs from GDSC, respectively. Then, the DL models were refined on HGCC, where the best source drug was identified. Finally, the DL model was validated on GSE232173. Using 2-step TL with pretraining on oxaliplatin was not only superior to those without TL and with 1-step TL but also better than 3 benchmark methods, including MGMT. The oxaliplatin-based TL improved the performance probably by increasing the weights of cell cycle-related genes, which relates to the TMZ response processes. Our findings support the potential of oxaliplatin being an alternative therapy for patients with GBM and TL facilitating drug repurposing research. We recommend that following our methodology, using mixed cancers and a related drug as the source and then fine-tuning the model with the target cancer and the target drug will enhance drug response prediction. |
| format | Article |
| id | doaj-art-09fe2beb7d2241e49ca32af6dcdd3ccc |
| institution | Kabale University |
| issn | 1177-9322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | SAGE Publishing |
| record_format | Article |
| series | Bioinformatics and Biology Insights |
| spelling | doaj-art-09fe2beb7d2241e49ca32af6dcdd3ccc2025-01-04T07:03:23ZengSAGE PublishingBioinformatics and Biology Insights1177-93222025-01-011910.1177/11779322241301507Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of GlioblastomaJie Ju0Ioannis Ntafoulis1Michelle Klein2Marcel JT Reinders3Martine Lamfers4Andrew P Stubbs5Yunlei Li6Department of Pathology & Clinical Bioinformatics, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Neurosurgery, Brain Tumor Center, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Neurosurgery, Brain Tumor Center, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsThe Delft Bioinformatics Lab, Delft University of Technology, Delft, The NetherlandsDepartment of Neurosurgery, Brain Tumor Center, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Pathology & Clinical Bioinformatics, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsDepartment of Pathology & Clinical Bioinformatics, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The NetherlandsWhile deep learning (DL) is used in patients’ outcome predictions, the insufficiency of patient samples limits the accuracy. In this study, we investigated how transfer learning (TL) alleviates the small sample size problem. A 2-step TL framework was constructed for a difficult task: predicting the response of the drug temozolomide (TMZ) in glioblastoma (GBM) cell cultures. The GBM is aggressive, and most patients do not benefit from the only approved chemotherapeutic agent TMZ. O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status is the only biomarker for TMZ responsiveness but has shown limited predictive power. The 2-step TL framework was built on 3 datasets: (1) the subset of the Genomics of Drug Sensitivity in Cancer (GDSC) dataset, including miscellaneous cell cultures treated by TMZ, cyclophosphamide, bortezomib, and oxaliplatin, as the source dataset; (2) the Human Glioblastoma Cell Culture (HGCC) dataset, for fine-tuning; and (3) a small target dataset GSE232173, for validation. The latter two included specifically TMZ-treated GBM cell cultures. The DL models were pretrained on the cell cultures treated by each of the 4 drugs from GDSC, respectively. Then, the DL models were refined on HGCC, where the best source drug was identified. Finally, the DL model was validated on GSE232173. Using 2-step TL with pretraining on oxaliplatin was not only superior to those without TL and with 1-step TL but also better than 3 benchmark methods, including MGMT. The oxaliplatin-based TL improved the performance probably by increasing the weights of cell cycle-related genes, which relates to the TMZ response processes. Our findings support the potential of oxaliplatin being an alternative therapy for patients with GBM and TL facilitating drug repurposing research. We recommend that following our methodology, using mixed cancers and a related drug as the source and then fine-tuning the model with the target cancer and the target drug will enhance drug response prediction.https://doi.org/10.1177/11779322241301507 |
| spellingShingle | Jie Ju Ioannis Ntafoulis Michelle Klein Marcel JT Reinders Martine Lamfers Andrew P Stubbs Yunlei Li Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma Bioinformatics and Biology Insights |
| title | Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma |
| title_full | Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma |
| title_fullStr | Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma |
| title_full_unstemmed | Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma |
| title_short | Two-Step Transfer Learning Improves Deep Learning–Based Drug Response Prediction in Small Datasets: A Case Study of Glioblastoma |
| title_sort | two step transfer learning improves deep learning based drug response prediction in small datasets a case study of glioblastoma |
| url | https://doi.org/10.1177/11779322241301507 |
| work_keys_str_mv | AT jieju twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT ioannisntafoulis twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT michelleklein twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT marceljtreinders twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT martinelamfers twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT andrewpstubbs twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma AT yunleili twosteptransferlearningimprovesdeeplearningbaseddrugresponsepredictioninsmalldatasetsacasestudyofglioblastoma |