SMSSVD: SubMatrix Selection Singular Value Decomposition

Henningsson, Rasmus; Fontes, Magnus

doi:10.1093/bioinformatics/bty566

Cited by 5 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, a dimension estimate of the data was obtained by generating a Talus plot (Fig. 1, bottom left panel, and Supplementary Material), after which noise reduction was performed by SubMatrix Selection Singular Value Decomposition (SMSSVD) (Henningsson and Fontes 2019). SMSSVD is ideal for situations where complex data containing a very large number of variables have signals spread out over different (possibly overlapping) subsets of variables, with the goal of recovering all signals that can be detected, rather than only the strongest one.…”

Section: Resultsmentioning

confidence: 99%

“…SMSSVD (Henningsson and Fontes 2019) is a parameter-free dimension reduction technique designed for the reconstruction of multiple overlaid low-rank signals from a data matrix, corrupted by noise. It is ideal for exploratory analysis of complex data, where different signals are spread out over different (possibly overlapping) subsets of variables, by limiting the influence of noise in variables that are not contributing to the signal.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

DISSEQT—DIStribution-based modeling of SEQuence space Time dynamics†

et al. 2019

View full text Add to dashboard Cite

Rapidly evolving microbes are a challenge to model because of the volatile, complex, and dynamic nature of their populations. We developed the DISSEQT pipeline (DIStribution-based SEQuence space Time dynamics) for analyzing, visualizing, and predicting the evolution of heterogeneous biological populations in multidimensional genetic space, suited for population-based modeling of deep sequencing and high-throughput data. The pipeline is openly available on GitHub (https://github.com/rasmushenningsson/DISSEQT.jl, accessed 23 June 2019) and Synapse (https://www.synapse.org/#!Synapse: syn11425758, accessed 23 June 2019), covering the entire workflow from read alignment to visualization of results. Our pipeline is centered around robust dimension and model reduction algorithms for analysis of genotypic data with additional capabilities for including phenotypic features to explore dynamic genotype–phenotype maps. We illustrate its utility and capacity with examples from evolving RNA virus populations, which present one of the highest degrees of genetic heterogeneity within a given population found in nature. Using our pipeline, we empirically reconstruct the evolutionary trajectories of evolving populations in sequence space and genotype–phenotype fitness landscapes. We show that while sequence space is vastly multidimensional, the relevant genetic space of evolving microbial populations is of intrinsically low dimension. In addition, evolutionary trajectories of these populations can be faithfully monitored to identify the key minority genotypes contributing most to evolution. Finally, we show that empirical fitness landscapes, when reconstructed to include minority variants, can predict phenotype from genotype with high accuracy.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

DISSEQT—DIStribution-based modeling of SEQuence space Time dynamics†

et al. 2019

View full text Add to dashboard Cite

show abstract

“…PLS has long been common in genomics [10,11,33] , though it remains uncommon in statistics and machine learning, and its theoretical properties are poorly understood. Other recent PCA-based approaches for genetics, though not directly applicable for prediction are SMSSVD [25] and ESPCA [39].…”

Section: Recent Related Workmentioning

confidence: 99%

Sufficient principal component regression for pattern discovery in transcriptomic data

Ding,

Zentner,

McDonald

2021

Preprint

View full text Add to dashboard Cite

Methods for global measurement of transcript abundance such as microarrays and RNAseq generate datasets in which the number of measured features far exceeds the number of observations. Extracting biologically meaningful and experimentally tractable insights from such data therefore requires high-dimensional prediction. Existing sparse linear approaches to this challenge have been stunningly successful, but some important issues remain. These methods can fail to select the correct features, predict poorly relative to non-sparse alternatives, or ignore any unknown grouping structures for the features. We propose a method called SuffPCR that yields improved predictions in high-dimensional tasks including regression and classification, especially in the typical context of omics with correlated features. SuffPCR first estimates sparse principal components and then estimates a linear model on the recovered subspace. Because the estimated subspace is sparse in the features, the resulting predictions will depend on only a small subset of genes. SuffPCR works well on a variety of simulated and experimental transcriptomic data, performing nearly optimally when the model assumptions are satisfied. We also demonstrate near-optimal theoretical guarantees.

show abstract

“…Second, a dimension estimate of the data was obtained by generating a Talus plot ( Figure 1, bottom left panel, and Supplementary Methods), after which noise reduction was performed by SubMatrix Selection Singular Value Decomposition (SMSSVD) 25 . SMSSVD is ideal for situations where complex data containing a very large number of variables have signals spread out over different (possibly overlapping) subsets of variables, with the goal of recovering all signals that can be detected, rather than only the strongest one.…”

Section: Overview Of the Disseqt Pipelinementioning

confidence: 99%

“…SubMatrix Selection Singular Value Decomposition (SMSSVD) 25 is a parameterfree dimension reduction technique designed for the reconstruction of multiple overlaid low-rank signals from a data matrix, corrupted by noise. It is ideal for exploratory analysis of complex data, where different signals are spread out over different (possibly overlapping) subsets of variables, by limiting the influence of noise in variables that are not contributing to the signal.…”

Section: Smssvdmentioning

confidence: 99%

DISSEQT - DIStribution based modeling of SEQuence space Time dynamics

Henningsson¹,

Moratorio²,

Bordería³

et al. 2018

Preprint

View full text Add to dashboard Cite

Rapidly evolving microbes are a challenge to model because of the volatile, complex and dynamic nature of their populations. We developed the DISSEQT pipeline (DIStribution-based SEQuence space Time dynamics) for analyzing, visualizing and predicting the evolution of heterogeneous biological populations in multidimensional genetic space, suited for population-based modeling of deep sequencing and high-throughput data. DISSEQT is openly available on GitHub (https://github.com/ rasmushenningsson/DISSEQT.jl) and Synapse (https://www.synapse. org/#!Synapse:syn11425758), covering the entire workflow from read alignment to visualization of results. DISSEQT is centered around robust dimension and model reduction algorithms for analysis of genotypic data with additional capabilities for including phenotypic features to explore dynamic genotype-phenotype maps. We illustrate its utility and capacity with examples from evolving RNA virus populations, which present on of the highest degrees of population heterogeneity found in nature. Using DISSEQT, we empirically reconstruct the evolutionary trajectories of evolving populations in sequence space and genotype-phenotype fitness landscapes. We show that while sequence space is vastly multidimensional, the relevant genetic space of evolving microbial populations is of intrinsically low dimension. In addition, evolutionary trajectories of these populations can be faithfully monitored to identify the key minority genotypes contributing most to evolution. Finally, we show that empirical fitness landscapes, when reconstructed to include minority variants, can predict phenotype from genotype with high accuracy.

show abstract

SMSSVD: SubMatrix Selection Singular Value Decomposition

Abstract: Supplementary Materials are available at Bioinformatics online.

Cited by 5 publications

References 17 publications

DISSEQT—DIStribution-based modeling of SEQuence space Time dynamics†

DISSEQT—DIStribution-based modeling of SEQuence space Time dynamics†

Sufficient principal component regression for pattern discovery in transcriptomic data

DISSEQT - DIStribution based modeling of SEQuence space Time dynamics

Contact Info

Product

Resources

About