Dataset Information

Supervised non-negative matrix factorization methods for MALDI imaging applications.

ABSTRACT: MOTIVATION:Non-negative matrix factorization (NMF) is a common tool for obtaining low-rank approximations of non-negative data matrices and has been widely used in machine learning, e.g. for supporting feature extraction in high-dimensional classification tasks. In its classical form, NMF is an unsupervised method, i.e. the class labels of the training data are not used when computing the NMF. However, incorporating the classification labels into the NMF algorithms allows to specifically guide them toward the extraction of data patterns relevant for discriminating the respective classes. This approach is particularly suited for the analysis of mass spectrometry imaging (MSI) data in clinical applications, such as tumor typing and classification, which are among the most challenging tasks in pathology. Thus, we investigate algorithms for extracting tumor-specific spectral patterns from MSI data by NMF methods. RESULTS:In this article, we incorporate a priori class labels into the NMF cost functional by adding appropriate supervised penalty terms. Numerical experiments on a MALDI imaging dataset confirm that the novel supervised NMF methods lead to significantly better classification accuracy and stability as compared with other standard approaches. AVAILABILITY AND IMPLEMENTATON:https://gitlab.informatik.uni-bremen.de/digipath/Supervised_NMF_Methods_for_MALDI.git. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.

SUBMITTER: Leuschner J

PROVIDER: S-EPMC6546133 | biostudies-literature | 2019 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Supervised non-negative matrix factorization methods for MALDI imaging applications.

Leuschner Johannes J Schmidt Maximilian M Fernsel Pascal P Lachmund Delf D Boskamp Tobias T Maass Peter P

Bioinformatics (Oxford, England) 20190601 11

<h4>Motivation</h4>Non-negative matrix factorization (NMF) is a common tool for obtaining low-rank approximations of non-negative data matrices and has been widely used in machine learning, e.g. for supporting feature extraction in high-dimensional classification tasks. In its classical form, NMF is an unsupervised method, i.e. the class labels of the training data are not used when computing the NMF. However, incorporating the classification labels into the NMF algorithms allows to specifically ...[more]

PMID: 30395171

Similar Datasets

Project description:MotivationUnderstanding the underlying mutational processes of cancer patients has been a long-standing goal in the community and promises to provide new insights that could improve cancer diagnoses and treatments. Mutational signatures are summaries of the mutational processes, and improving the derivation of mutational signatures can yield new discoveries previously obscured by technical and biological confounders. Results from existing mutational signature extraction methods depend on the size of available patient cohort and solely focus on the analysis of mutation count data without considering the exploitation of metadata.ResultsHere we present a supervised method that utilizes cancer type as metadata to extract more distinctive signatures. More specifically, we use a negative binomial non-negative matrix factorization and add a support vector machine loss. We show that mutational signatures extracted by our proposed method have a lower reconstruction error and are designed to be more predictive of cancer type than those generated by unsupervised methods. This design reduces the need for elaborate post-processing strategies in order to recover most of the known signatures unlike the existing unsupervised signature extraction methods. Signatures extracted by a supervised model used in conjunction with cancer-type labels are also more robust, especially when using small and potentially cancer-type limited patient cohorts. Finally, we adapted our model such that molecular features can be utilized to derive an according mutational signature. We used APOBEC expression and MUTYH mutation status to demonstrate the possibilities that arise from this ability. We conclude that our method, which exploits available metadata, improves the quality of mutational signatures as well as helps derive more interpretable representations.Availability and implementationhttps://github.com/ratschlab/SNBNMF-mutsig-public.Supplementary informationSupplementary data are available at Bioinformatics online.

Project description:Background:Matrix factorization is a well established pattern discovery tool that has seen numerous applications in biomedical data analytics, such as gene expression co-clustering, patient stratification, and gene-disease association mining. Matrix factorization learns a latent data model that takes a data matrix and transforms it into a latent feature space enabling generalization, noise removal and feature discovery. However, factorization algorithms are numerically intensive, and hence there is a pressing challenge to scale current algorithms to work with large datasets. Our focus in this paper is matrix tri-factorization, a popular method that is not limited by the assumption of standard matrix factorization about data residing in one latent space. Matrix tri-factorization solves this by inferring a separate latent space for each dimension in a data matrix, and a latent mapping of interactions between the inferred spaces, making the approach particularly suitable for biomedical data mining. Results:We developed a block-wise approach for latent factor learning in matrix tri-factorization. The approach partitions a data matrix into disjoint submatrices that are treated independently and fed into a parallel factorization system. An appealing property of the proposed approach is its mathematical equivalence with serial matrix tri-factorization. In a study on large biomedical datasets we show that our approach scales well on multi-processor and multi-GPU architectures. On a four-GPU system we demonstrate that our approach can be more than 100-times faster than its single-processor counterpart. Conclusions:A general approach for scaling non-negative matrix tri-factorization is proposed. The approach is especially useful parallel matrix factorization implemented in a multi-GPU environment. We expect the new approach will be useful in emerging procedures for latent factor analysis, notably for data integration, where many large data matrices need to be collectively factorized.

Project description:RationaleNon-negative matrix factorization (NMF) has been used extensively for the analysis of mass spectrometry imaging (MSI) data, visualizing simultaneously the spatial and spectral distributions present in a slice of tissue. The statistical framework offers two related NMF methods: probabilistic latent semantic analysis (PLSA) and latent Dirichlet allocation (LDA), which is a generative model. This work offers a mathematical comparison between NMF, PLSA, and LDA, and includes a detailed evaluation of Kullback-Leibler NMF (KL-NMF) for MSI for the first time. We will inspect the results for MSI data analysis as these different mathematical approaches impose different characteristics on the data and the resulting decomposition.MethodsThe four methods (NMF, KL-NMF, PLSA, and LDA) are compared on seven different samples: three originated from mice pancreas and four from human-lymph-node tissues, all obtained using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS).ResultsWhere matrix factorization methods are often used for the analysis of MSI data, we find that each method has different implications on the exactness and interpretability of the results. We have discovered promising results using KL-NMF, which has only rarely been used for MSI so far, improving both NMF and PLSA, and have shown that the hitherto stated equivalent KL-NMF and PLSA algorithms do differ in the case of MSI data analysis. LDA, assumed to be the better method in the field of text mining, is shown to be outperformed by PLSA in the setting of MALDI-MSI. Additionally, the molecular results of the human-lymph-node data have been thoroughly analyzed for better assessment of the methods under investigation.ConclusionsWe present an in-depth comparison of multiple NMF-related factorization methods for MSI. We aim to provide fellow researchers in the field of MSI a clear understanding of the mathematical implications using each of these analytical techniques, which might affect the exactness and interpretation of the results.

Dataset Information

Supervised non-negative matrix factorization methods for MALDI imaging applications.

Publications

Supervised non-negative matrix factorization methods for MALDI imaging applications.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets