-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathREADME.Rmd
88 lines (59 loc) · 3.81 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
---
title: "`r read.dcf('DESCRIPTION', fields = 'Package')[1]`"
author: "`r rworkflows::use_badges()`"
date: "<h4>README updated: <i>`r format( Sys.Date(), '%b-%d-%Y')`</i></h4>"
output:
github_document
---
<!-- To modify Package/Title/Description/Authors fields, edit the DESCRIPTION file -->
```{r, echo=FALSE, include=FALSE}
pkg <- read.dcf("DESCRIPTION", fields = "Package")[1]
title <- read.dcf("DESCRIPTION", fields = "Title")[1]
description <- gsub("\n"," ",read.dcf("DESCRIPTION", fields = "Description")[1])
URL <- read.dcf('DESCRIPTION', fields = 'URL')[1]
owner <- strsplit(URL,"/")[[1]][4]
repo <-strsplit(URL,"/")[[1]][5]
```
## Introduction
Many genes have been associated with diseases
Multi-Scale Target Explorer (`MSTExplorer`) systematically identifies, prioritises, and visualises cell-type-specific gene therapy targets across the phenome.
Core functionalities include:
**1. Conducting phenotype x cell type genetic association tests at scale**
- The [Human Phenotype Ontology](https://hpo.jax.org/) (integrated with gene annotations from [OMIM](https://omim.org/) / [DECIPHER](https://www.deciphergenomics.org/) / [ORPHANET](https://www.orpha.net/)) is used as the source of phenotype gene signatures. Each gene-phenotype associated is given a continuous score that approximates the current strength of evidence for the association (using data derived from [GenCC](https://thegencc.org/)).
- Whole-body scRNA-seq atlases from humans (across multiple developmental stages) are used as a data-driven source of cell type-specific gene markers.
- The underlying association tests are designed for both speed and accuracy using memory-efficient data structures, and a highly parallelisable implementation of Generalised Linear Regression (GLM). For example, associations for all pairwise combinations of >11k phenotypes x >200 cell types (>2,200,000 associations) can be in <30 minutes on a Macbook laptop with 10 CPU cores).
**2. Inferring multi-scale causal graphs of disease**
`MSTExplorer` allows users to easily infer and construct multi-scale causal graphs of Diseases (blue nodes) -> Phenotypes (purple nodes) -> Cell types (orange nodes) -> Genes (yellow nodes).
{height=400px}
[See here for more example networks.](https://github.com/neurogenomics/rare_disease_celltyping/tree/299abe0ccd00644bc2f05a1389704fe196a3e868/manuscript/_manuscript/img/fig-therapy-examples-supp).
**3. Prioritising cell-type-specific gene therapy targets**
`MSTExplorer` also provides a comprehensive and customisable pipeline that can be run via a single function (`prioritise_targets()`) to produce the most promising cell-type-specific gene therapy targets across the phenome.
## Installation
Within R:
```R
if(!require("BiocManager")) install.packages("BiocManager")
BiocManager::install("`r paste(owner,repo,sep='/')`")
library(`r pkg`)
```
## Documentation
#### [Website](https://`r owner`.github.io/`r pkg`)
#### [Get started](https://`r owner`.github.io/`r pkg`/articles/`r pkg`)
#### [Docker/Singularity Container](https://`r owner`.github.io/`r pkg`/articles/docker.html)
## Citation
If you use ``r pkg``, please cite:
<!-- Modify this by editing the file: inst/CITATION -->
> `r utils::citation(pkg)$textVersion`
## Contact
### [Neurogenomics Lab](https://www.neurogenomics.co.uk)
UK Dementia Research Institute
Department of Brain Sciences
Faculty of Medicine
Imperial College London
[GitHub](https://github.com/neurogenomics)
## Session Info
<details>
```{r Session Info}
utils::sessionInfo()
```
</details>
<hr>