Periodic Reporting for period 1 - CODer (The molecular basis and genetic control of local gene co-expression and its impact in human disease)
Período documentado: 2020-06-01 hasta 2022-05-31
The genetic makeup intrinsic to each person shapes their particular traits, disease susceptibilities and treatment effectiveness, making understanding the functional impact of mutations one of the most pursued challenges in genetics research. The action “The molecular basis of local gene co-expression and its impact on human disease” attempted to address several ongoing questions in genetics pertaining to how genes are interconnected and linked to disease. Up until recently, the field has focused on associating individual genes to disease, yet, thousands of genes in the human genome are active in the same cell and thus should be taken together in analysis. Considering multiple genes together could help explain disease aetiology as well as disease comorbidity, driven by mutations affecting multiple co-active genes. Yet, the co-activity of genes that are nearby each other is currently understudied. Moreover, gene activity varies across human tissues which provides a challenge to its analysis.
This project aimed to tackle these challenges and address several questions:
- Determine the extent of nearby gene co-activity across tissues and its regulation by genetic variants
- Discover the regulatory mechanisms and evolutionary reasons underlying gene co-activity
- Evaluate the potential impact of nearby gene co-activity in human disease
A key step of this project consisted in determining the co-activity of genes across all available human tissues and cell types. For this, the largest publicly available dataset of gene activity measurements across human tissues was exploited to identify maps of nearby gene co-activity. Producing these co-activity maps allowed the investigation of the molecular and evolutionary reasons for gene co-activity as well as identifying mutations predicted to affect the activity of not only one but multiple nearby genes. The project revealed thousands of mutations that are able to affect the activity of multiple nearby genes - due to the shared regulation of gene activity - and provided first evidence that such mutations may affect a higher number of human diseases and traits than mutations only affecting one gene. The project shed new light to the interpretation of genetic mutation results and provided the community with maps of gene co-activity and the elements involved in their co-regulation.
This project aimed to tackle these challenges and address several questions:
- Determine the extent of nearby gene co-activity across tissues and its regulation by genetic variants
- Discover the regulatory mechanisms and evolutionary reasons underlying gene co-activity
- Evaluate the potential impact of nearby gene co-activity in human disease
A key step of this project consisted in determining the co-activity of genes across all available human tissues and cell types. For this, the largest publicly available dataset of gene activity measurements across human tissues was exploited to identify maps of nearby gene co-activity. Producing these co-activity maps allowed the investigation of the molecular and evolutionary reasons for gene co-activity as well as identifying mutations predicted to affect the activity of not only one but multiple nearby genes. The project revealed thousands of mutations that are able to affect the activity of multiple nearby genes - due to the shared regulation of gene activity - and provided first evidence that such mutations may affect a higher number of human diseases and traits than mutations only affecting one gene. The project shed new light to the interpretation of genetic mutation results and provided the community with maps of gene co-activity and the elements involved in their co-regulation.
Work was conducted through 3 work packages (WPs). WP1 consisted in developing an analysis framework to identify gene co-activity across dozens of human tissues. For this, the Fellow deployed large-scale statistical analysis using large computing clusters to produce millions of computations. Moreover, thousands of mutations linked to gene activity were identified using previously established methods. WP2 sought to exploit the dataset developed in WP1 to obtain biologically-relevant insights, in particular with the aim of understanding the molecular and evolutionary reasons of gene co-activity. WP3 involved the use of state-of-the-art datasets of single cell measurements (high-detail molecular data) to allow the identification of nearby gene co-activity per individual, something not previously attempted in the field. Several pertinent findings were obtained, including:
- The commonalities of gene co-activity patterns across different human tissues
- The role of shared regulatory elements in gene co-activity
- The effect of mutations in gene co-activity and disease
Overview of the results and their dissemination
Besides novel biological insights, the action produced the following concrete results:
- A novel framework to perform nearby gene co-activity analysis
- A comprehensive map of gene co-activity and associated mutations for each human tissue
- Gene and enhancer co-activity maps derived from single cell data in two human cell types
These results were disseminated in the following manner:
- Two scientific publications, one published in a peer-reviewed scientific journal and another available as a preprint and currently in revision in a peer-reviewed journal
- Ribeiro, D.M. Rubinacci, S., Ramisch, A. et al. The molecular basis, genetic control and pleiotropic effects of local gene co-expression. Nat Commun 12, 4842 (2021). https://doi.org/10.1038/s41467-021-25129-x
- Ribeiro, D.M. Ziyani, C., Delaneau, O. Shared regulation and functional relevance of local gene co-expression revealed by single cell analysis. bioRxiv (2021). https://doi.org/10.1101/2021.12.14.472573
- The novel framework was made available though commonly used code repositories (github)
- Maps of gene co-activity were made available to the scientific community through a dedicated public database
- Results were further disseminated through 7 conferences, including through plenary talks at the major international conferences in the field, and 6 other seminars
- The commonalities of gene co-activity patterns across different human tissues
- The role of shared regulatory elements in gene co-activity
- The effect of mutations in gene co-activity and disease
Overview of the results and their dissemination
Besides novel biological insights, the action produced the following concrete results:
- A novel framework to perform nearby gene co-activity analysis
- A comprehensive map of gene co-activity and associated mutations for each human tissue
- Gene and enhancer co-activity maps derived from single cell data in two human cell types
These results were disseminated in the following manner:
- Two scientific publications, one published in a peer-reviewed scientific journal and another available as a preprint and currently in revision in a peer-reviewed journal
- Ribeiro, D.M. Rubinacci, S., Ramisch, A. et al. The molecular basis, genetic control and pleiotropic effects of local gene co-expression. Nat Commun 12, 4842 (2021). https://doi.org/10.1038/s41467-021-25129-x
- Ribeiro, D.M. Ziyani, C., Delaneau, O. Shared regulation and functional relevance of local gene co-expression revealed by single cell analysis. bioRxiv (2021). https://doi.org/10.1101/2021.12.14.472573
- The novel framework was made available though commonly used code repositories (github)
- Maps of gene co-activity were made available to the scientific community through a dedicated public database
- Results were further disseminated through 7 conferences, including through plenary talks at the major international conferences in the field, and 6 other seminars
During the course of the action the Fellow accomplished several major achievements in gene regulation and genomics research:
A. Established the prevalence of nearby gene co-activity across dozens human tissues, which goes well beyond the state-of-the-art in both the depth of the tissues as well as the statistical validity of the discoveries (Ribeiro et al. 2021 Nature Comm. 12, 4842).
B. Proposed a novel statistical method to determine nearby gene co-activity based on (i) robust measurements across hundreds of individuals, (ii) accounting for confounding factors, (iii) use of permutation approach to measure statistical significance, all of which consist of improvements over previous research work.
C. Determined the relative importance of several molecular features (regulatory elements, chromatin contacts, expression variation) in nearby gene co-activity as well as the functional relatedness of co-active genes.
D. Ascertained the pervasiveness of genetic mutations affecting the activity of multiple genes and measured their potential pleiotropic consequences for human traits and diseases (UK biobank phenotypes). This provided first evidence that, by affecting several genes, shared mutations could have stronger implications in disease.
E. Demonstrated the feasibility of using gene activity measurements in single cells across multiple individuals for predicting nearby co-active genes. This constitutes a new usage of single cell data which unlocks novel analysis such as the comparison of co-activity events across individuals (Ribeiro et al. 2021 bioRxiv DOI 10.1101/2021.12.14.472573).
F. Evaluated the use of cutting-edge multimodal single cell technologies in identifying links between genes and their regulatory elements in the genome, allowing for interactions of regulatory elements to multiple genes.
Until now, few studies had focused on the gene co-activity that occurs between nearby genes. The study of co-activity between nearby genes is particularly relevant as these could occur through distinct molecular mechanisms (e.g. other than transcription factor regulation) and evolutionary reasons. Large-scale studies to explore gene activity across the whole human genome and dozens of tissues as the one performed here are essential for obtaining a global picture of nearby gene co-activity and its regulation and disease-impact. The work developed during the action pioneered the dissection of nearby gene co-activity using single cell analysis and the datasets disseminated during this action are expected to be widely exploited by scientists in the field and will be kept available in the coming years.
A. Established the prevalence of nearby gene co-activity across dozens human tissues, which goes well beyond the state-of-the-art in both the depth of the tissues as well as the statistical validity of the discoveries (Ribeiro et al. 2021 Nature Comm. 12, 4842).
B. Proposed a novel statistical method to determine nearby gene co-activity based on (i) robust measurements across hundreds of individuals, (ii) accounting for confounding factors, (iii) use of permutation approach to measure statistical significance, all of which consist of improvements over previous research work.
C. Determined the relative importance of several molecular features (regulatory elements, chromatin contacts, expression variation) in nearby gene co-activity as well as the functional relatedness of co-active genes.
D. Ascertained the pervasiveness of genetic mutations affecting the activity of multiple genes and measured their potential pleiotropic consequences for human traits and diseases (UK biobank phenotypes). This provided first evidence that, by affecting several genes, shared mutations could have stronger implications in disease.
E. Demonstrated the feasibility of using gene activity measurements in single cells across multiple individuals for predicting nearby co-active genes. This constitutes a new usage of single cell data which unlocks novel analysis such as the comparison of co-activity events across individuals (Ribeiro et al. 2021 bioRxiv DOI 10.1101/2021.12.14.472573).
F. Evaluated the use of cutting-edge multimodal single cell technologies in identifying links between genes and their regulatory elements in the genome, allowing for interactions of regulatory elements to multiple genes.
Until now, few studies had focused on the gene co-activity that occurs between nearby genes. The study of co-activity between nearby genes is particularly relevant as these could occur through distinct molecular mechanisms (e.g. other than transcription factor regulation) and evolutionary reasons. Large-scale studies to explore gene activity across the whole human genome and dozens of tissues as the one performed here are essential for obtaining a global picture of nearby gene co-activity and its regulation and disease-impact. The work developed during the action pioneered the dissection of nearby gene co-activity using single cell analysis and the datasets disseminated during this action are expected to be widely exploited by scientists in the field and will be kept available in the coming years.