-
Notifications
You must be signed in to change notification settings - Fork 1
/
README.Rmd
70 lines (46 loc) · 3.67 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
---
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%"
)
```
# JAM: Joint Analysis of Marginal summary statistics
<!-- badges: start -->
[](https://github.com/lailylajiang/hJAM)
<!-- badges: end -->
<!-- CRAN badges: start -->
[](https://CRAN.R-project.org/package=hJAM)[](https://CRAN.R-project.org/package=hJAM)
<!-- CRAN badges: end -->
## About
In this package, we provide three `JAM`-family methods that utilize GWAS summary statistics to perform post-GWAS analysis, including Mendelian Randomization, TWAS, and multi-population fine-mapping.
## hJAM
hJAM (`hJAM`) is a hierarchical model which unifies the framework of Mendelian Randomization and Transcriptome-wide association studies. The hJAM-Egger (`hJAM_egger`) is a natural extension on hJAM that uses an intercept term to account for the pleiotropy effect of the SNPs on the outcome.
Additionally, we provide implementations to construct the weight matrix `A` in hJAM. `JAM_A` converts the marginal summary statistics into conditional ones by using the correlation matrix of the SNPs from a reference data. `susieJAM_A` selects SNPs for one intermediate using the marginal summary statistics from GWAS or other study summary data.
## SHA-JAM
SHA-JAM (`SHAJAM`) is a scalable version of `hJAM` that handles high-dimensional intermediates. SHA-JAM performs model selection from highly correlated intermediates through SuSiE: Sum of Single Effect Model (`SHAJAM`) or elastic-net (`EN.hJAM`).
## mJAM
mJAM is for multi-population fine-mapping using GWAS summary statistics and credible set construction. We provide two implementations of mJAM: one through SuSiE (`mJAM_SuSiE`) and another one through forward selection (`mJAM_Forward`). `mJAM_Forward` also provides the flexiblity of constructing credible sets for using user-defined index SNPs. A tutorial to get started with mJAM can be found [here](https://jiayi-s.github.io/more_on_mJAM/).
## Citing this work
* **hJAM**: Jiang, L., Xu, S., Mancuso, N., Newcombe, P. J., & Conti, D. V. (2021). A Hierarchical Approach Using Marginal Summary Statistics for Multiple Intermediates in a Mendelian Randomization or Transcriptome Analysis. *American journal of epidemiology*, 190(6), 1148–1158. https://doi.org/10.1093/aje/kwaa287
* **SHA-JAM**: Jiang, L., Conti, D.V. SHA-JAM: A Scalable Hierarchical Approach to Joint Analysis for Marginal Summary Statistics with Omics Data. (In preparation).
* **mJAM**: Shen, J., Jiang, L., Wang, K., Wang, A., Chen, F., Newcombe, P.J., Haiman, C.A., & Conti, D.V. Fine-Mapping and Credible Set Construction using a Multi-population Joint Analysis of Marginal Summary Statistics from Genome-wide Association Studies. ([Preprint](https://www.biorxiv.org/content/10.1101/2022.12.22.521659v1) available)
## Quick start with hJAM package
You can install the published version of hJAM from CRAN with:
``` r
install.packages("hJAM")
```
Currently we are working on improving `hJAM` package and adding genome-wide implementation of mJAM. If you want to use the most updated version of this package, we recommend installing the development version from [GitHub](https://github.com/USCbiostats/hJAM) with:
``` r
if (!require("devtools")) { install.packages("devtools") } else {}
devtools::install_github("USCbiostats/hJAM")
```
## Session info
```{r session info}
sessionInfo()
```