-
Notifications
You must be signed in to change notification settings - Fork 2
/
README.Rmd
174 lines (131 loc) · 8.53 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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
---
output: github_document
---
<!-- README.md is generated from README.Rmd. Please edit that file -->
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# TREG <a href="http://research.libd.org/TREG/"><img src="man/figures/logo.png" align="right" height="139" alt="TREG website" /></a>
<!-- badges: start -->
[](https://lifecycle.r-lib.org/articles/stages.html#stable)
[](https://bioconductor.org/checkResults/release/bioc-LATEST/TREG)
[](https://bioconductor.org/checkResults/devel/bioc-LATEST/TREG)
[](http://bioconductor.org/packages/stats/bioc/TREG/)
[](https://support.bioconductor.org/tag/TREG)
[](https://bioconductor.org/packages/release/bioc/html/TREG.html#since)
[](http://bioconductor.org/checkResults/devel/bioc-LATEST/TREG/)
[](https://bioconductor.org/packages/release/bioc/html/TREG.html#since)
[](https://codecov.io/gh/LieberInstitute/TREG?branch=devel)
[](https://github.com/LieberInstitute/TREG/actions)
[](https://github.com/LieberInstitute/TREG/issues)
[](https://github.com/LieberInstitute/TREG/pulls)
[](https://zenodo.org/badge/latestdoi/391101988)
<!-- badges: end -->
The goal of `TREG` is to help find candidate **Total RNA Expression Genes (TREGs)**
in single nucleus (or single cell) RNA-seq data.
_**Note**: TREG is pronounced as a single word and fully capitalized, unlike [Regulatory T cells](https://en.wikipedia.org/wiki/Regulatory_T_cell), which are known as "Tregs" (pronounced "T-regs"). The work described here is unrelated to regulatory T cells._
### Why are TREGs useful?
The expression of a TREG is proportional to the the overall RNA expression in a
cell. This relationship can be used to estimate total RNA content in cells in
assays where only a few genes can be measured, such as single-molecule
fluorescent in situ hybridization (smFISH).
In a smFISH experiment the number of TREG puncta can be used to infer the total
RNA expression of the cell.
The motivation of this work is to collect data via smFISH in to help build better
deconvolution algorithms. But may be many other application for TREGs in
experimental design!
<p align="center">
{width=50%}
</p>
### What makes a gene a good TREG?
1. The gene must have **non-zero expression in most cells** across different tissue
and cell types.
2. A TREG should also be expressed at a constant level in respect to other genes
across different cell types or have **high rank invariance**.
3. Be **measurable as a continuous metric** in the experimental assay, for example
have a dynamic range of puncta when observed in RNAscope. This will need to be
considered for the candidate TREGs, and may need to be validated experimentally.
<p align="center">
{width=30%}
</p>
### How to find candidate TREGs with `TREG`
<p align="center">
{width=100%}
</p>
1. **Filter for low Proportion Zero genes snRNA-seq dataset:** This is
facilitated with the functions `get_prop_zero()` and `filter_prop_zero()`.
snRNA-seq data is notoriously sparse, these functions enrich for genes with more
universal expression.
2. **Evaluate genes for Rank Invariance** The nuclei are grouped only
by cell type. Within each cell type, the mean expression for each
gene is ranked, the result is a vector (length is the number of
genes), using the function `rank_group()`. Then the expression of each gene is
ranked for each nucleus,the result is a matrix (the number of nuclei x number
of genes), using the function `rank_cells()`.Then the absolute difference
between the rank of each nucleus and the mean expression is found, from here
the mean of the differences for each gene is calculated, then ranked.
These steps are repeated for each group, the result is a matrix of ranks, (number of cell
types x number of genes). From here the sum of the ranks for each
gene are reversed ranked, so there is one final value for each gene,
the “Rank Invariance” The genes with the highest rank-invariance are
considered good candidates as TREGs. This is calculated with `rank_invariance_express()`.
**This full process is implemented by: `rank_invariance_express()`.**
## Installation instructions
Get the latest stable `R` release from [CRAN](http://cran.r-project.org/). Then install `TREG` using from [Bioconductor](http://bioconductor.org/) the following code:
```{r 'install', eval = FALSE}
if (!requireNamespace("BiocManager", quietly = TRUE)) {
install.packages("BiocManager")
}
BiocManager::install("TREG")
```
And the development version from [GitHub](https://github.com/LieberInstitute/TREG) with:
```{r 'install_dev', eval = FALSE}
BiocManager::install("LieberInstitute/TREG")
```
## Example
```{r `libraries`, message = FALSE, warning=FALSE}
## Load packages
library("TREG")
```
### Proportion Zero Filtering
A TREG gene should be expressed in almost every cell. The set of
genes should be filtered by maximum Proportion Zero within a groups of cells.
```{r calc_prop_zero, eval = requireNamespace('TREG')}
## Calculate Proportion Zero in groups defined by a column in colData
(prop_zero <- get_prop_zero(sce = sce_zero_test, group_col = "cellType"))
## Get list of genes that pass the max Proportion Zero filter
(filtered_genes <- filter_prop_zero(prop_zero, cutoff = 0.9))
## Filter sce object to this list of genes
sce_filter <- sce_zero_test[filtered_genes, ]
```
### Evaluate RI for Filtered SCE Data
The genes with the highest Rank Invariance are considered good candidates as TREGs.
In this example the gene *g0* would be the strongest candidate TREG.
```{r run_RI, eval = requireNamespace('TREG')}
## Get the Rank Invariance value for each gene
## The highest values are the best TREG candidates
ri <- rank_invariance_express(sce_filter)
sort(ri, decreasing = TRUE)
```
## Citation
Below is the citation output from using `citation('TREG')` in R. Please
run this yourself to check for any updates on how to cite __TREG__.
```{r 'citation', eval = requireNamespace('TREG')}
print(citation("TREG"), bibtex = TRUE)
```
Please note that the `TREG` was only made possible thanks to many other R and bioinformatics software authors, which are cited either in the vignettes and/or the paper(s) describing this package.
## Code of Conduct
Please note that the `TREG` project is released with a [Contributor Code of Conduct](http://bioconductor.org/about/code-of-conduct/). By contributing to this project, you agree to abide by its terms.
## Development tools
* Continuous code testing is possible thanks to [GitHub actions](https://www.tidyverse.org/blog/2020/04/usethis-1-6-0/) through `r BiocStyle::CRANpkg('usethis')`, `r BiocStyle::CRANpkg('remotes')`, and `r BiocStyle::CRANpkg('rcmdcheck')` customized to use [Bioconductor's docker containers](https://www.bioconductor.org/help/docker/) and `r BiocStyle::Biocpkg('BiocCheck')`.
* Code coverage assessment is possible thanks to [codecov](https://codecov.io/gh) and `r BiocStyle::CRANpkg('covr')`.
* The [documentation website](http://LieberInstitute.github.io/TREG) is automatically updated thanks to `r BiocStyle::CRANpkg('pkgdown')`.
* The code is styled automatically thanks to `r BiocStyle::CRANpkg('styler')`.
* The documentation is formatted thanks to `r BiocStyle::CRANpkg('devtools')` and `r BiocStyle::CRANpkg('roxygen2')`.
For more details, check the `dev` directory.
This package was developed using `r BiocStyle::Biocpkg('biocthis')`.