web-rMKL: a web server for dimensionality reduction and sample clustering of multi-view data based on unsupervised multiple kernel learning

Röder, Benedict; Kersten, Nicolas; Herr, Marius; Speicher, Nora K.; Pfeifer, Nico

doi:10.1093/nar/gkz422

Cited by 6 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the supervised learning problem, both approaches yielded the same result in terms of F1-score and AUROC, but ESCAPED outperformed the randomized encoding based approach in terms of execution time. In the unsupervised learning case, we replicated the state-of-the-art experiments conducted by Röder et al (2019) in a privacy preserving way without sacrificing performance. This indicates that ESCAPED enables making privacy preserving multi-omics dimensionality reduction and clustering whereas it was not possible to compute the required kernel matrices with the randomized encoding based approach, which is one of the fastest competitors, due to the excessive memory usage.…”

Section: Discussionmentioning

confidence: 99%

“…Head and Neck Squamous Cell Carcinoma Dataset: We aim to perform the privacy preserving multi-omics dimensionality reduction and clustering on the TCGA data for head and neck squamous cell carcinoma (HNSC) (Network et al 2015) to stratify patients into clinically meaningful subgroups. Therefore, we replicate a recent state-of-the-art study (Röder et al 2019) in a privacy-preserving setting, obtaining the data from the authors. The data consists of 465 patients with their gene expression (IlluminaHiSeq), DNA methylation (Methy-lation450k), copy number variation (gistic2), and miRNA expression (IlluminaHiSeq) data.…”

Section: Datamentioning

confidence: 99%

“…The method was recently evaluated as the best method on a large benchmark study comparing many different methods (Rappoport and Shamir 2018). Later, Röder et al (2019) published the online version of the method called web-rMKL. In that study, one of their use cases is the identification of subgroups of HNSC.…”

Section: Clustering Of Hnsc Cancer Patientsmentioning

confidence: 99%

See 2 more Smart Citations

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

Ünal¹,

Akgün²,

Pfeifer³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

To train sophisticated machine learning models one usually needs many training samples. Especially in healthcare settings these samples can be very expensive, meaning that one institution alone usually does not have enough on its own. Merging privacy-sensitive data from different sources is usually restricted by data security and data protection measures. This can lead to approaches that reduce data quality by putting noise onto the variables (e.g., in differential privacy) or omitting certain values (e.g., for k-anonymity). Other measures based on cryptographic methods can lead to very timeconsuming computations, which is especially problematic for larger multi-omics data. We address this problem by introducing ESCAPED, which stands for Efficient SeCure And PrivatE Dot product framework, enabling the computation of the dot product of vectors from multiple sources on a thirdparty, which later trains kernel-based machine learning algorithms, while neither sacrificing privacy nor adding noise. We evaluated our framework on drug resistance prediction for HIV-infected people and multi-omics dimensionality reduction and clustering problems in precision medicine. In terms of execution time, our framework significantly outperforms the best-fitting existing approaches without sacrificing the performance of the algorithm. Even though we only show the benefit for kernel-based algorithms, our framework can open up new research opportunities for further machine learning models that require the dot product of vectors from multiple sources.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Datamentioning

confidence: 99%

See 1 more Smart Citation

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

Ünal¹,

Akgün²,

Pfeifer³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Head and Neck Squamous Cell Carcinoma Dataset: We aim to perform the privacy preserving multi-omics dimensionality reduction and clustering on the TCGA data for head and neck squamous cell carcinoma (HNSC) (Cancer Genome Atlas Network et al 2015) to stratify patients into clinically meaningful subgroups. Therefore, we replicate a recent stateof-the-art study (Röder et al 2019) in a privacy-preserving setting, obtaining the data from the authors. The data consists of 465 patients with their gene expression (IlluminaHiSeq), DNA methylation (Methylation450k), copy number variation (gistic2), and miRNA expression (IlluminaHiSeq) data.…”

Section: Datamentioning

confidence: 99%

“…The method was recently evaluated as the best method in a large benchmark study that compared many different methods (Rappoport and Shamir 2018). Later, Röder et al (2019) published the online version of the method called web-rMKL.…”

Section: Clustering Of Hnsc Cancer Patientsmentioning

confidence: 99%

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

Ünal

Akgün

Pfeifer

2021

AAAI

Self Cite

View full text Add to dashboard Cite

Training sophisticated machine learning models usually requires many training samples. Especially in healthcare settings these samples can be very expensive, meaning that one institution alone usually does not have enough. Merging privacy-sensitive data from different sources is usually restricted by data security and data protection measures. This can lead to approaches that reduce data quality by putting noise onto the variables (e.g., in epsilon-differential privacy) or omitting certain values (e.g., for k-anonymity). Other measures based on cryptographic methods can lead to very time-consuming computations, which is especially problematic for larger multi-omics data. We address this problem by introducing ESCAPED, which stands for Efficient SeCure And PrivatE Dot product framework. ESCAPED enables the computation of the dot product of vectors from multiple sources on a third-party, which later trains kernel-based machine learning algorithms, while neither sacrificing privacy nor adding noise. We have evaluated our framework on drug resistance prediction for HIV-infected people and multi-omics dimensionality reduction and clustering problems in precision medicine. In terms of execution time, our framework significantly outperforms the best-fitting existing approaches without sacrificing the performance of the algorithm. Even though we only present the benefit for kernel-based algorithms, our framework can open up new research opportunities for further machine learning models that require the dot product of vectors from multiple sources.

show abstract

Multi-Omics Data Mining Techniques: Algorithms and Software

Tang

Liu

Gong

2023

Methodologies of Multi-Omics Data Integration and Data Mining

View full text Add to dashboard Cite

web-rMKL: a web server for dimensionality reduction and sample clustering of multi-view data based on unsupervised multiple kernel learning

Cited by 6 publications

References 12 publications

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

ESCAPED: Efficient Secure and Private Dot Product Framework for Kernel-based Machine Learning Algorithms with Applications in Healthcare

Multi-Omics Data Mining Techniques: Algorithms and Software

Contact Info

Product

Resources

About