Cargando…
Cross-species analysis of enhancer logic using deep learning
Deciphering the genomic regulatory code of enhancers is a key challenge in biology because this code underlies cellular identity. A better understanding of how enhancers work will improve the interpretation of noncoding genome variation and empower the generation of cell type–specific drivers for ge...
Autores principales: | , , , , , , , , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Cold Spring Harbor Laboratory Press
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7706731/ https://www.ncbi.nlm.nih.gov/pubmed/32732264 http://dx.doi.org/10.1101/gr.260844.120 |
_version_ | 1783617210684538880 |
---|---|
author | Minnoye, Liesbeth Taskiran, Ibrahim Ihsan Mauduit, David Fazio, Maurizio Van Aerschot, Linde Hulselmans, Gert Christiaens, Valerie Makhzami, Samira Seltenhammer, Monika Karras, Panagiotis Primot, Aline Cadieu, Edouard van Rooijen, Ellen Marine, Jean-Christophe Egidy, Giorgia Ghanem, Ghanem-Elias Zon, Leonard Wouters, Jasper Aerts, Stein |
author_facet | Minnoye, Liesbeth Taskiran, Ibrahim Ihsan Mauduit, David Fazio, Maurizio Van Aerschot, Linde Hulselmans, Gert Christiaens, Valerie Makhzami, Samira Seltenhammer, Monika Karras, Panagiotis Primot, Aline Cadieu, Edouard van Rooijen, Ellen Marine, Jean-Christophe Egidy, Giorgia Ghanem, Ghanem-Elias Zon, Leonard Wouters, Jasper Aerts, Stein |
author_sort | Minnoye, Liesbeth |
collection | PubMed |
description | Deciphering the genomic regulatory code of enhancers is a key challenge in biology because this code underlies cellular identity. A better understanding of how enhancers work will improve the interpretation of noncoding genome variation and empower the generation of cell type–specific drivers for gene therapy. Here, we explore the combination of deep learning and cross-species chromatin accessibility profiling to build explainable enhancer models. We apply this strategy to decipher the enhancer code in melanoma, a relevant case study owing to the presence of distinct melanoma cell states. We trained and validated a deep learning model, called DeepMEL, using chromatin accessibility data of 26 melanoma samples across six different species. We show the accuracy of DeepMEL predictions on the CAGI5 challenge, where it significantly outperforms existing models on the melanoma enhancer of IRF4. Next, we exploit DeepMEL to analyze enhancer architectures and identify accurate transcription factor binding sites for the core regulatory complexes in the two different melanoma states, with distinct roles for each transcription factor, in terms of nucleosome displacement or enhancer activation. Finally, DeepMEL identifies orthologous enhancers across distantly related species, where sequence alignment fails, and the model highlights specific nucleotide substitutions that underlie enhancer turnover. DeepMEL can be used from the Kipoi database to predict and optimize candidate enhancers and to prioritize enhancer mutations. In addition, our computational strategy can be applied to other cancer or normal cell types. |
format | Online Article Text |
id | pubmed-7706731 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | Cold Spring Harbor Laboratory Press |
record_format | MEDLINE/PubMed |
spelling | pubmed-77067312020-12-09 Cross-species analysis of enhancer logic using deep learning Minnoye, Liesbeth Taskiran, Ibrahim Ihsan Mauduit, David Fazio, Maurizio Van Aerschot, Linde Hulselmans, Gert Christiaens, Valerie Makhzami, Samira Seltenhammer, Monika Karras, Panagiotis Primot, Aline Cadieu, Edouard van Rooijen, Ellen Marine, Jean-Christophe Egidy, Giorgia Ghanem, Ghanem-Elias Zon, Leonard Wouters, Jasper Aerts, Stein Genome Res Method Deciphering the genomic regulatory code of enhancers is a key challenge in biology because this code underlies cellular identity. A better understanding of how enhancers work will improve the interpretation of noncoding genome variation and empower the generation of cell type–specific drivers for gene therapy. Here, we explore the combination of deep learning and cross-species chromatin accessibility profiling to build explainable enhancer models. We apply this strategy to decipher the enhancer code in melanoma, a relevant case study owing to the presence of distinct melanoma cell states. We trained and validated a deep learning model, called DeepMEL, using chromatin accessibility data of 26 melanoma samples across six different species. We show the accuracy of DeepMEL predictions on the CAGI5 challenge, where it significantly outperforms existing models on the melanoma enhancer of IRF4. Next, we exploit DeepMEL to analyze enhancer architectures and identify accurate transcription factor binding sites for the core regulatory complexes in the two different melanoma states, with distinct roles for each transcription factor, in terms of nucleosome displacement or enhancer activation. Finally, DeepMEL identifies orthologous enhancers across distantly related species, where sequence alignment fails, and the model highlights specific nucleotide substitutions that underlie enhancer turnover. DeepMEL can be used from the Kipoi database to predict and optimize candidate enhancers and to prioritize enhancer mutations. In addition, our computational strategy can be applied to other cancer or normal cell types. Cold Spring Harbor Laboratory Press 2020-12 /pmc/articles/PMC7706731/ /pubmed/32732264 http://dx.doi.org/10.1101/gr.260844.120 Text en © 2020 Minnoye et al.; Published by Cold Spring Harbor Laboratory Press http://creativecommons.org/licenses/by/4.0/ This article, published in Genome Research, is available under a Creative Commons License (Attribution 4.0 International), as described at http://creativecommons.org/licenses/by/4.0/. |
spellingShingle | Method Minnoye, Liesbeth Taskiran, Ibrahim Ihsan Mauduit, David Fazio, Maurizio Van Aerschot, Linde Hulselmans, Gert Christiaens, Valerie Makhzami, Samira Seltenhammer, Monika Karras, Panagiotis Primot, Aline Cadieu, Edouard van Rooijen, Ellen Marine, Jean-Christophe Egidy, Giorgia Ghanem, Ghanem-Elias Zon, Leonard Wouters, Jasper Aerts, Stein Cross-species analysis of enhancer logic using deep learning |
title | Cross-species analysis of enhancer logic using deep learning |
title_full | Cross-species analysis of enhancer logic using deep learning |
title_fullStr | Cross-species analysis of enhancer logic using deep learning |
title_full_unstemmed | Cross-species analysis of enhancer logic using deep learning |
title_short | Cross-species analysis of enhancer logic using deep learning |
title_sort | cross-species analysis of enhancer logic using deep learning |
topic | Method |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7706731/ https://www.ncbi.nlm.nih.gov/pubmed/32732264 http://dx.doi.org/10.1101/gr.260844.120 |
work_keys_str_mv | AT minnoyeliesbeth crossspeciesanalysisofenhancerlogicusingdeeplearning AT taskiranibrahimihsan crossspeciesanalysisofenhancerlogicusingdeeplearning AT mauduitdavid crossspeciesanalysisofenhancerlogicusingdeeplearning AT faziomaurizio crossspeciesanalysisofenhancerlogicusingdeeplearning AT vanaerschotlinde crossspeciesanalysisofenhancerlogicusingdeeplearning AT hulselmansgert crossspeciesanalysisofenhancerlogicusingdeeplearning AT christiaensvalerie crossspeciesanalysisofenhancerlogicusingdeeplearning AT makhzamisamira crossspeciesanalysisofenhancerlogicusingdeeplearning AT seltenhammermonika crossspeciesanalysisofenhancerlogicusingdeeplearning AT karraspanagiotis crossspeciesanalysisofenhancerlogicusingdeeplearning AT primotaline crossspeciesanalysisofenhancerlogicusingdeeplearning AT cadieuedouard crossspeciesanalysisofenhancerlogicusingdeeplearning AT vanrooijenellen crossspeciesanalysisofenhancerlogicusingdeeplearning AT marinejeanchristophe crossspeciesanalysisofenhancerlogicusingdeeplearning AT egidygiorgia crossspeciesanalysisofenhancerlogicusingdeeplearning AT ghanemghanemelias crossspeciesanalysisofenhancerlogicusingdeeplearning AT zonleonard crossspeciesanalysisofenhancerlogicusingdeeplearning AT woutersjasper crossspeciesanalysisofenhancerlogicusingdeeplearning AT aertsstein crossspeciesanalysisofenhancerlogicusingdeeplearning |