Towards in silico CLIP-seq

Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Towards in silico CLIP-seq : predicting protein-RNA interaction via sequence-to-signal learning. / Horlacher, Marc; Wagner, Nils; Moyon, Lambert; Kuret, Klara; Goedert, Nicolas; Salvatore, Marco; Ule, Jernej; Gagneur, Julien; Winther, Ole; Marsico, Annalisa.

In: Genome Biology, Vol. 24, No. 1, 180, 2023.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Horlacher, M, Wagner, N, Moyon, L, Kuret, K, Goedert, N, Salvatore, M, Ule, J, Gagneur, J, Winther, O & Marsico, A 2023, 'Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning', Genome Biology, vol. 24, no. 1, 180. https://doi.org/10.1186/s13059-023-03015-7

APA

Horlacher, M., Wagner, N., Moyon, L., Kuret, K., Goedert, N., Salvatore, M., Ule, J., Gagneur, J., Winther, O., & Marsico, A. (2023). Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning. Genome Biology, 24(1), [180]. https://doi.org/10.1186/s13059-023-03015-7

Vancouver

Horlacher M, Wagner N, Moyon L, Kuret K, Goedert N, Salvatore M et al. Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning. Genome Biology. 2023;24(1). 180. https://doi.org/10.1186/s13059-023-03015-7

Author

Horlacher, Marc ; Wagner, Nils ; Moyon, Lambert ; Kuret, Klara ; Goedert, Nicolas ; Salvatore, Marco ; Ule, Jernej ; Gagneur, Julien ; Winther, Ole ; Marsico, Annalisa. / Towards in silico CLIP-seq : predicting protein-RNA interaction via sequence-to-signal learning. In: Genome Biology. 2023 ; Vol. 24, No. 1.

Bibtex

@article{11664553313a4e81985312574840c5ec,

title = "Towards in silico CLIP-seq: predicting protein-RNA interaction via sequence-to-signal learning",

abstract = "We present RBPNet, a novel deep learning method, which predicts CLIP-seq crosslink count distribution from RNA sequence at single-nucleotide resolution. By training on up to a million regions, RBPNet achieves high generalization on eCLIP, iCLIP and miCLIP assays, outperforming state-of-the-art classifiers. RBPNet performs bias correction by modeling the raw signal as a mixture of the protein-specific and background signal. Through model interrogation via Integrated Gradients, RBPNet identifies predictive sub-sequences that correspond to known and novel binding motifs and enables variant-impact scoring via in silico mutagenesis. Together, RBPNet improves imputation of protein-RNA interactions, as well as mechanistic interpretation of predictions.",

keywords = "CLIP-seq, Computational biology, Deep learning, Protein-RNA interaction",

author = "Marc Horlacher and Nils Wagner and Lambert Moyon and Klara Kuret and Nicolas Goedert and Marco Salvatore and Jernej Ule and Julien Gagneur and Ole Winther and Annalisa Marsico",

note = "Publisher Copyright: {\textcopyright} 2023, BioMed Central Ltd., part of Springer Nature.",

year = "2023",

doi = "10.1186/s13059-023-03015-7",

language = "English",

volume = "24",

journal = "Genome Biology (Online Edition)",

issn = "1474-7596",

publisher = "BioMed Central Ltd.",

number = "1",

}

RIS

TY - JOUR

T1 - Towards in silico CLIP-seq

T2 - predicting protein-RNA interaction via sequence-to-signal learning

AU - Horlacher, Marc

AU - Wagner, Nils

AU - Moyon, Lambert

AU - Kuret, Klara

AU - Goedert, Nicolas

AU - Salvatore, Marco

AU - Ule, Jernej

AU - Gagneur, Julien

AU - Winther, Ole

AU - Marsico, Annalisa

PY - 2023

Y1 - 2023

N2 - We present RBPNet, a novel deep learning method, which predicts CLIP-seq crosslink count distribution from RNA sequence at single-nucleotide resolution. By training on up to a million regions, RBPNet achieves high generalization on eCLIP, iCLIP and miCLIP assays, outperforming state-of-the-art classifiers. RBPNet performs bias correction by modeling the raw signal as a mixture of the protein-specific and background signal. Through model interrogation via Integrated Gradients, RBPNet identifies predictive sub-sequences that correspond to known and novel binding motifs and enables variant-impact scoring via in silico mutagenesis. Together, RBPNet improves imputation of protein-RNA interactions, as well as mechanistic interpretation of predictions.

AB - We present RBPNet, a novel deep learning method, which predicts CLIP-seq crosslink count distribution from RNA sequence at single-nucleotide resolution. By training on up to a million regions, RBPNet achieves high generalization on eCLIP, iCLIP and miCLIP assays, outperforming state-of-the-art classifiers. RBPNet performs bias correction by modeling the raw signal as a mixture of the protein-specific and background signal. Through model interrogation via Integrated Gradients, RBPNet identifies predictive sub-sequences that correspond to known and novel binding motifs and enables variant-impact scoring via in silico mutagenesis. Together, RBPNet improves imputation of protein-RNA interactions, as well as mechanistic interpretation of predictions.

KW - CLIP-seq

KW - Computational biology

KW - Deep learning

KW - Protein-RNA interaction

U2 - 10.1186/s13059-023-03015-7

DO - 10.1186/s13059-023-03015-7

M3 - Journal article

C2 - 37542318

AN - SCOPUS:85166598318

VL - 24

JO - Genome Biology (Online Edition)

JF - Genome Biology (Online Edition)

SN - 1474-7596

IS - 1

M1 - 180

ER -

ID: 362744991

Department of Biology