caoluolong/xomicsenrich

Integration omics summary data and enrichment analysis

XomicsEnrich

Integration omics summary data and enrichment analysis

Useage:

Run the "XomicsEnrich/main/XomicsEnrich.m" in the matlab, and youcan refer to a GUI like the below figure:

1. Select the input file.

A text file could be read by matlab(.txt, .xls, ...), and contains at least 2 column, the first column is a list of brain regions, the following column(s) is corresponding region's eigenvalue (trait).For example:

Here we collected 5 types of brain atlas totally. All brain region IDs of different atlas were united coded and recorded in "./rawdata/brain_atlas/".

2. Select brain atlas and parameters.

Brain Atlas: DK68_aparcaseg; 500.aparc; power264PNC; AAL1_ROI_MNI_V4; AAL3v1_1mm
Sphere: left; ap2left; whole; map2whole
Interpolate: no; nearest; gauss

3. Input PPI parameters.

PPI database: String full version11.5; String full version12.0; String physical version11.5; String physical version12.0;
PPI neighber: a integer k (default=1), the k-th neighbour of a gene in the PPI.
PPI quantile: a decimal number j between 0 and 1 (default=0.5), the threshold of connection edge in the PPI.

4. Input partial least squares regression parameters.

PLS dimension: the dimention of PLS analysis. Typically the first 5 dim could explain enough of the trait variance.
PLS permutation: the number of permutation times in analysis. Typically 1000 times permutation enables the PLS analysis have a constant dim-signaficance result.
PLS bootstrap: the number of bootstrap times in analysis. Typically 3000 times bootstrap enables the PLS analysis have a constant factor-weight result.

5. The output path.

Output path: the result of analysis will be stored at this path automately.

6. Load and Analysis.

Output results.

1. a "Run.log" file records he parameters and time of this analysis program.
2. a .mat file contains gene expression matrix and PET map of biomoleculars.
3. a .fig file depicts the variance proportation of trait explained by genes/PETs on PLS analysis.
4. a .mat file contains gene/PET weights on trait. This depict a gene could impact the trait through which biomolecules(PETs).
5. a .txt file contains: all genes' association with trait by totaly 10 integration methods (rows * columns).
6. a .txt file contains: all genes' association with trait by 6 of the 10 methods.
7. a .txt file contains: all genes' association with trait by addPPI method only.

Posthoc enrichment analysis with R code.

The enrichment analysis can refer to "./utils/pathway_enrichment.R" (run in R session), and you can do GO term, KEGG pathway, or celltype enrichent analysis and plot results of interest.