Information content differentiates enhancers from silencers in mouse photoreceptors

Friedman, Ryan Z.; Granas, David M; Myers, Connie A.; Corbo, Joseph C.; Cohen, Barak A.; White, Michael A.

doi:10.7554/elife.67403

Cited by 24 publications

(114 citation statements)

References 89 publications

(217 reference statements)

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…This SVM-based method has been previously used to highlight specific sequence features in the mouse retinal epigenome and to predict retinal-reporter expression post-hoc. [31][32][33][34] Together these previous studies and our current work demonstrate the potential of this approach to characterize human retinal CRE sequences for the identification of crucial features and their variants. While this approach can be applied to many types of sequencing data, the use of general chromatin accessibility via ATAC-seq allows the model to incorporate the sequence features of diverse regulatory elements in a less biased approach than using more specific ChIP-seq data.…”

Section: Discussionsupporting

confidence: 74%

See 1 more Smart Citation

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal Cis-Regulatory Elements

VandenBosch

Luu

Timms

et al. 2021

Preprint

View full text Add to dashboard Cite

PurposePrior studies demonstrate the significance of specific cis-regulatory variants in retinal disease, however determining the functional impact of regulatory variants remains a major challenge. In this study, we utilize a machine learning approach, trained on epigenomic data from the adult human retina, to systematically quantify the predicted impact of cis-regulatory variants.MethodsWe used human retinal DNA accessibility data (ATAC-seq) to determine a set of 18.9k high-confidence putative cis-regulatory elements. 80% of these elements were used to train a machine learning model utilizing a gapped k-mer support vector machine-based approach. In silico saturation mutagenesis and variant scoring was applied to predict the functional impact of all potential single nucleotide variants within cis-regulatory elements. Impact scores were tested in a 20% hold-out dataset and compared to allele population frequency, phylogenetic conservation, transcription factor (TF) binding motifs, and existing massively parallel reporter assay (MPRA) data.ResultsWe generated a model that distinguishes between human retinal regulatory elements and negative test sequences with 95% accuracy. Among a hold-out test set of 3.7k human retinal CREs, all possible single nucleotide variants (SNVs) were scored. Variants with negative impact scores correlated with reduced population allele frequency, higher phylogenetic conservation of the reference allele, disruption of predicted TF binding motifs, and massively-parallel reporter expression.ConclusionsWe demonstrated the utility of human retinal epigenomic data to train a machine learning model for the purpose of predicting the impact of non-coding regulatory sequence variants. Our model accurately scored sequences and predicted putative transcription factor binding motifs. This approach has the potential to expedite the characterization of pathogenic non-coding sequence variants in the context of unexplained retinal disease.

show abstract

Section: Discussionsupporting

confidence: 74%

“…[28][29][30] This GKM-SVM approach has been applied successfully to predict sequence values in the context of specific mouse retinal enhancers. [31][32][33][34] However, to date, it has not been applied across a wider set of human retinal epigenomic data to perform a comprehensive prediction of CRE variant impact scores.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal Cis-Regulatory Elements

VandenBosch

Luu

Timms

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…This SVM-based method has been previously used to highlight specific sequence features in the mouse retinal epigenome and to predict retinal reporter expression post hoc. 34 – 37 Together these previous studies and our current work demonstrate the potential of this approach to characterize human retinal CRE sequences for the identification of crucial features and their variants. Although this approach can be applied to many types of sequencing data, the use of general chromatin accessibility via ATAC-seq allows the model to incorporate the sequence features of diverse regulatory elements in a less biased approach than using more specific ChIP-seq data.…”

Section: Discussionsupporting

confidence: 70%

“… 31 – 33 This GKM-SVM approach has been applied successfully to predict sequence values in the context of specific mouse retinal enhancers. 34 – 37 However, to date, it has not been applied across a wider set of human retinal epigenomic data to perform a comprehensive prediction of CRE variant impact scores.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal cis-Regulatory Elements

VandenBosch

Luu

Timms

et al. 2022

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

Purpose Prior studies have demonstrated the significance of specific cis -regulatory variants in retinal disease; however, determining the functional impact of regulatory variants remains a major challenge. In this study, we utilized a machine learning approach, trained on epigenomic data from the adult human retina, to systematically quantify the predicted impact of cis -regulatory variants. Methods We used human retinal DNA accessibility data (ATAC-seq) to determine a set of 18.9k high-confidence, putative cis -regulatory elements. Eighty percent of these elements were used to train a machine learning model utilizing a gapped k -mer support vector machine–based approach. In silico saturation mutagenesis and variant scoring was applied to predict the functional impact of all potential single nucleotide variants within cis -regulatory elements. Impact scores were tested in a 20% hold-out dataset and compared to allele population frequency, phylogenetic conservation, transcription factor (TF) binding motifs, and existing massively parallel reporter assay data. Results We generated a model that distinguishes between human retinal regulatory elements and negative test sequences with 95% accuracy. Among a hold-out test set of 3.7k human retinal CREs, all possible single nucleotide variants were scored. Variants with negative impact scores correlated with higher phylogenetic conservation of the reference allele, disruption of predicted TF binding motifs, and massively parallel reporter expression. Conclusions We demonstrated the utility of human retinal epigenomic data to train a machine learning model for the purpose of predicting the impact of non-coding regulatory sequence variants. Our model accurately scored sequences and predicted putative transcription factor binding motifs. This approach has the potential to expedite the characterization of pathogenic non-coding sequence variants in the context of unexplained retinal disease. Translational Relevance This workflow and resulting dataset serve as a promising genomic tool to facilitate the clinical prioritization of functionally disruptive non-coding mutations in the retina.

show abstract

“…Therefore, the direction of the bifunctionality (whether the gene gets activated or whether it becomes repressed) might be tuned by the DBD's motif within an enhancer/silencer. Evidence for this hypothesis has been shown for the bifunctionality of the homeodomain TF CRX where observing repression, in addition to activation, depends on the number of CRX binding motifs in a synthetic context or the presence of other TF binding motifs in a genomic context 49,50 . The functional difference between a bifunctional protein's silencer or enhancer sequence might simply be explained by the bifunctionality of the effector domain and the ratio of recruited activating and repressing complexes, where more binding motifs will lead to repression.…”

Section: Discussionmentioning

confidence: 99%

Large-scale mapping and systematic mutagenesis of human transcriptional effector domains

DelRosso

Tycko

Suzuki

et al. 2022

Preprint

View full text Add to dashboard Cite

Human gene expression is regulated by over two thousand transcription factors and chromatin regulators. Effector domains within these proteins can activate or repress transcription. However, for many of these regulators we do not know what type of transcriptional effector domains they contain, their location in the protein, their activation and repression strengths, and the amino acids that are necessary for their functions. Here, we systematically measure the transcriptional effector activity of >100,000 protein fragments (each 80 amino acids long) tiling across most chromatin regulators and transcription factors in human cells (2,047 proteins). By testing the effect they have when recruited at reporter genes, we annotate 307 new activation domains and 592 new repression domains, a ~5-fold increase over the number of previously annotated effectors. Complementary rational mutagenesis and deletion scans across all the effector domains reveal aromatic and/or leucine residues interspersed with acidic, proline, serine, and/or glutamine residues are necessary for activation domain activity. Additionally, the majority of repression domain sequences contain either sites for SUMOylation, short interaction motifs for recruiting co-repressors, or are structured binding domains for recruiting other repressive proteins. Surprisingly, we discover bifunctional domains that can both activate and repress and can dynamically split a cell population into high- and low-expression subpopulations. Our systematic annotation and characterization of transcriptional effector domains provides a rich resource for understanding the function of human transcription factors and chromatin regulators, engineering compact tools for controlling gene expression, and refining predictive computational models of effector domain function.

show abstract

Information content differentiates enhancers from silencers in mouse photoreceptors

Cited by 24 publications

References 89 publications

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal Cis-Regulatory Elements

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal Cis-Regulatory Elements

Machine Learning Prediction of Non-Coding Variant Impact in Human Retinal cis-Regulatory Elements

Large-scale mapping and systematic mutagenesis of human transcriptional effector domains

Contact Info

Product

Resources

About