Results on Dataset-1.
<div><p>Early and accurate cancer detection is crucial for effective treatment, prognosis, and the advancement of precision medicine. Analyzing omics data is vital in cancer research. While using a single type of omics data provides a limited perspective, integrating multiple omics modal...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , , , , |
| منشور في: |
2025
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1852015734076473344 |
|---|---|
| author | Ma Yinghua (22446698) |
| author2 | Ahmad Khan (6067526) Yang Heng (10181637) Fiaz Gul Khan (22446701) Afnan Aldhahri (22446704) Iftikhar Ahmed Khan (5679773) |
| author2_role | author author author author author |
| author_facet | Ma Yinghua (22446698) Ahmad Khan (6067526) Yang Heng (10181637) Fiaz Gul Khan (22446701) Afnan Aldhahri (22446704) Iftikhar Ahmed Khan (5679773) |
| author_role | author |
| dc.creator.none.fl_str_mv | Ma Yinghua (22446698) Ahmad Khan (6067526) Yang Heng (10181637) Fiaz Gul Khan (22446701) Afnan Aldhahri (22446704) Iftikhar Ahmed Khan (5679773) |
| dc.date.none.fl_str_mv | 2025-10-16T17:41:47Z |
| dc.identifier.none.fl_str_mv | 10.1371/journal.pone.0333134.t002 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Results_on_Dataset-1_/30378518 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Pharmacology Science Policy Biological Sciences not elsewhere classified xlink "> early preserving global information triplet margin loss composite loss function proposed model combines omics data provides omics data integration downstream classification tasks analyzing omics data preventing information loss enhance precision medicine end contrastive manner accurate cancer detection contrastive adversarial encoder omics data information loss adversarial loss precision medicine model achieved classification accuracy weighted combination vision transformer utilizing self single type results show paper proposes often results limited perspective latent space findings demonstrate feature redundancy f1 score eliminating redundancy effective treatment comprehensive understanding class classifications cancer research |
| dc.title.none.fl_str_mv | Results on Dataset-1. |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | <div><p>Early and accurate cancer detection is crucial for effective treatment, prognosis, and the advancement of precision medicine. Analyzing omics data is vital in cancer research. While using a single type of omics data provides a limited perspective, integrating multiple omics modalities allows for a more comprehensive understanding of cancer. Current deep models struggle to achieve efficient dimensionality reduction while preserving global information and integrating multi-omics data. This often results in feature redundancy or information loss, overlooking the synergies among different modalities. This paper proposes a contrastive adversarial encoder (CAEncoder) for multi-omics data integration to address this challenge. The proposed model combines a Vision Transformer (ViT) and a CycleGAN, trained in an end-to-end contrastive manner. The ViT is the encoder, utilizing self-attention, while the CycleGAN employs adversarial learning to ensure more discriminative and invariant latent space embeddings. Contrastive adversarial training improves representation quality by preventing information loss, eliminating redundancy, and capturing the synergies among different omics modalities. To ensure contrastive adversarial training, a composite loss function is used, consisting of a weighted combination of Adversarial Loss (Hinge Loss), Cycle Consistency Loss, and Triplet Margin Loss. The Adversarial Loss and Cycle Consistency Loss provide feedback from the CycleGAN, ensuring effective adversarial learning. Meanwhile, the Triplet Margin Loss promotes contrastive learning by pulling similar samples together and pushing dissimilar samples apart in the latent space. The performance of the CAEncoder is evaluated on downstream classification tasks, including both binary and multi-class classifications of five different cancer types. The results show that the model achieved a classification accuracy of up to 93.33% and an F1 score of 92.81%, outperforming existing advanced models. These findings demonstrate the potential of our method to enhance precision medicine for cancer through improved multi-omics data integration.</p></div> |
| eu_rights_str_mv | openAccess |
| id | Manara_5cbc4d1f09dbb4feadabfbdcf2d0766a |
| identifier_str_mv | 10.1371/journal.pone.0333134.t002 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/30378518 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Results on Dataset-1.Ma Yinghua (22446698)Ahmad Khan (6067526)Yang Heng (10181637)Fiaz Gul Khan (22446701)Afnan Aldhahri (22446704)Iftikhar Ahmed Khan (5679773)PharmacologyScience PolicyBiological Sciences not elsewhere classifiedxlink "> earlypreserving global informationtriplet margin losscomposite loss functionproposed model combinesomics data providesomics data integrationdownstream classification tasksanalyzing omics datapreventing information lossenhance precision medicineend contrastive manneraccurate cancer detectioncontrastive adversarial encoderomics datainformation lossadversarial lossprecision medicinemodel achievedclassification accuracyweighted combinationvision transformerutilizing selfsingle typeresults showpaper proposesoften resultslimited perspectivelatent spacefindings demonstratefeature redundancyf1 scoreeliminating redundancyeffective treatmentcomprehensive understandingclass classificationscancer research<div><p>Early and accurate cancer detection is crucial for effective treatment, prognosis, and the advancement of precision medicine. Analyzing omics data is vital in cancer research. While using a single type of omics data provides a limited perspective, integrating multiple omics modalities allows for a more comprehensive understanding of cancer. Current deep models struggle to achieve efficient dimensionality reduction while preserving global information and integrating multi-omics data. This often results in feature redundancy or information loss, overlooking the synergies among different modalities. This paper proposes a contrastive adversarial encoder (CAEncoder) for multi-omics data integration to address this challenge. The proposed model combines a Vision Transformer (ViT) and a CycleGAN, trained in an end-to-end contrastive manner. The ViT is the encoder, utilizing self-attention, while the CycleGAN employs adversarial learning to ensure more discriminative and invariant latent space embeddings. Contrastive adversarial training improves representation quality by preventing information loss, eliminating redundancy, and capturing the synergies among different omics modalities. To ensure contrastive adversarial training, a composite loss function is used, consisting of a weighted combination of Adversarial Loss (Hinge Loss), Cycle Consistency Loss, and Triplet Margin Loss. The Adversarial Loss and Cycle Consistency Loss provide feedback from the CycleGAN, ensuring effective adversarial learning. Meanwhile, the Triplet Margin Loss promotes contrastive learning by pulling similar samples together and pushing dissimilar samples apart in the latent space. The performance of the CAEncoder is evaluated on downstream classification tasks, including both binary and multi-class classifications of five different cancer types. The results show that the model achieved a classification accuracy of up to 93.33% and an F1 score of 92.81%, outperforming existing advanced models. These findings demonstrate the potential of our method to enhance precision medicine for cancer through improved multi-omics data integration.</p></div>2025-10-16T17:41:47ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1371/journal.pone.0333134.t002https://figshare.com/articles/dataset/Results_on_Dataset-1_/30378518CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/303785182025-10-16T17:41:47Z |
| spellingShingle | Results on Dataset-1. Ma Yinghua (22446698) Pharmacology Science Policy Biological Sciences not elsewhere classified xlink "> early preserving global information triplet margin loss composite loss function proposed model combines omics data provides omics data integration downstream classification tasks analyzing omics data preventing information loss enhance precision medicine end contrastive manner accurate cancer detection contrastive adversarial encoder omics data information loss adversarial loss precision medicine model achieved classification accuracy weighted combination vision transformer utilizing self single type results show paper proposes often results limited perspective latent space findings demonstrate feature redundancy f1 score eliminating redundancy effective treatment comprehensive understanding class classifications cancer research |
| status_str | publishedVersion |
| title | Results on Dataset-1. |
| title_full | Results on Dataset-1. |
| title_fullStr | Results on Dataset-1. |
| title_full_unstemmed | Results on Dataset-1. |
| title_short | Results on Dataset-1. |
| title_sort | Results on Dataset-1. |
| topic | Pharmacology Science Policy Biological Sciences not elsewhere classified xlink "> early preserving global information triplet margin loss composite loss function proposed model combines omics data provides omics data integration downstream classification tasks analyzing omics data preventing information loss enhance precision medicine end contrastive manner accurate cancer detection contrastive adversarial encoder omics data information loss adversarial loss precision medicine model achieved classification accuracy weighted combination vision transformer utilizing self single type results show paper proposes often results limited perspective latent space findings demonstrate feature redundancy f1 score eliminating redundancy effective treatment comprehensive understanding class classifications cancer research |