---
title: 'famat'
author: "Emilie Secherre"

output:
    BiocStyle::html_document
vignette: >
    %\VignetteIndexEntry{famat}
    %\VignetteEngine{knitr::rmarkdown}
    %\VignetteEncoding{UTF-8}
---

```{r style, echo = FALSE, results = 'asis'}
BiocStyle::markdown()
```

# Introduction

The aim of `r Rpackage("famat")` is to allow users to determine functional 
links between metabolites and genes. These metabolites and genes lists may 
be related to a specific experiment/study, but `r Rpackage("famat")` only 
needs a gene symbols list and a Kegg Compound ids list. Using these lists, 
`r Rpackage("famat")` performs pathway enrichment analysis, direct interactions
between elements inside pathways extraction, GO terms enrichment analysis, 
calculation of user's elements centrality (number of direct interactions 
between an element and others inside a pathway) and extraction of information
related to user's elements. 

Functions available are: <br><ul>
    <li> path_enrich : pathways enrichment analysis </li>
    <li> interactions : direct interactions and centrality </li>
    <li> compl_data : GO terms enrichment analysis and user's elements data
    extraction </li>
    <li> rshiny : use of previous function's data in a shiny interface </li>
</ul>

# Installation

Run this command line to install `r Rpackage("famat")`.
```{r, eval=FALSE}
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("famat")
```

Then, load `r Rpackage("famat")` using library.
```{r load_famat, message=FALSE, warning=FALSE}
library(famat)
library(mgcv)
```

# path_enrich

This function uses the metabolite list and the gene list provided by user to 
perform pathway enrichment analysis. Metabolites ids need to be Kegg compound
ids, and genes ids need to be gene symbols. Three pathway databases are 
available: Kegg ("KEGG"), Wikipathways ("WP") and Reactome ("REAC").
```{r pathenrich}
data(genes)
data(meta)

listr=path_enrich("REAC", meta, genes)
```
Results are then stored into a list. This list must be used in "interactions"
function.
Pathways enrichment analysis is performed on genes using 
`r CRANpkg("gprofiler2")` and on metabolites using `r CRANpkg("MPINet")`.

# interactions

"Interactions" find all direct interactions between genes and metabolites of
user's lists in pathways obtained through pathways enrichment analysis, 
performed on KEGG, Reactome and Wikipathways pathways. So, this function
needs results of "path_enrich" function performed on all these databases.
Using direct interactions, centrality of a user's element inside a pathway
is calculated.
```{r interactions}
data(listk)
data(listr)
data(listw)

interactions_result = interactions(listk, listr, listw)
```
Results are then stored into a list. This list must be used in "compl_data"
function.
Direct interactions were collected from BioPax, KGML and GPML files parsed 
with `r Biocpkg("PaxtoolsR")`, `r Biocpkg("graphite")` and author's parsers.
"Interactions" just get interactions of enriched pathways from this direct 
interactions list.

# compl_data

This function complete information about elements and pathway obtained with
"path_enrich" and "interactions". A GO term enrichment analysis is performed
on genes, pathways obtained through pathways enrichment analysis are filtered
(they must contain at least 1/5 elements in user's lists or a direct 
interaction between user's elements) and a hierarchy parent-child is built 
with pathways and enriched GO terms. GO terms enrichment analysis is performed
using `r Biocpkg("clusterProfiler")`. Then, dataframes containing information
about elements, interactions and GO terms are created, with an heatmap showing
which user's elements are in which pathways.
```{r compldata, message = FALSE}
data(interactions_result)

compl_data_result <- compl_data(interactions_result)
```
Results are then stored into a list. This list must be used in "rshiny"
function.

# rshiny

All results obtained with the three previous functions can be visualized using
"rshiny" function. `r CRANpkg("shiny")` is a R package allowing to create
interfaces.

```{r, eval=FALSE}
data(compl_data_result)

rshiny(compl_data_result)
```
After using this command line, the `r CRANpkg("shiny")` interface appear.

![test](./inte.png)
Interface's tabs are: <br><ul>
    <li> Elements : Genes, metabolites, interactions, and information related 
    to these elements. Also, filters can be applied to get pathways/enriched
    GO terms containing (or not) only certain elements. </li>
    <li> Pathways : Heatmap showing pathways hierarchies and user's elements
    in these pathways. Filters can be applied to get pathways corresponding 
    to certain pathways or interactions types, or to filter genes by their type
    or enriched GO terms. Also, the heatmap can be colored using interactions 
    types or centrality. Finally, click on a heatmap's cell makes a pop up 
    appear, with information about the pathway selected. </li>
    <li> GO Molecular Function : Enriched GO MF terms hierarchies with 
    information about them. Click on a GO term makes a pop up appear with data
    about the GO term selected. </li>
    <li> GO Biological Process : Enriched GO BP terms hierarchies with 
    information about them. Click on a GO term makes a pop up appear with data
    about the GO term selected. </li>
    <li> History : Every time a filter is applied, new results are saved in 
    history. So, user can go back if a mistake was done. </li>
    <li> Elements not in pathways : some user's elements may not be in pathways
    obtained by pathways enrichment analysis and filtered by compl_data 
    function. So, they are printed in this tab to show they were taken into 
    account. </li>
</ul>
Finally, a "Reset" button was made to go back to the initial results.

# Conclusion: how to use famat

To conclude, `r Rpackage("famat")` has four important functions which have to
be used one after another:
```{r famat, eval=FALSE}
data(genes)
data(meta)

listk <- path_enrich("KEGG", meta, genes)
listr <- path_enrich("REAC", meta, genes)
listw <- path_enrich("WP", meta, genes)

interactions_result <- interactions(listk, listr, listw)

compl_data_result <- compl_data(interactions_result)

rshiny(compl_data_result)
```

# References

<ul>
    <li>Yanjun Xu, Chunquan Li and Xia Li (2013). MPINet: The package can 
    implement the network-based metabolite pathway identification of pathways..
    R package version 1.0. https://CRAN.R-project.org/package=MPINet</li>
    <li>Liis Kolberg and Uku Raudvere (2020). gprofiler2: Interface to the
    'g:Profiler' Toolset. R package version 0.2.0.
    https://CRAN.R-project.org/package=gprofiler2 </li>
    <li>Luna, A., Babur, O., Aksoy, A. B, Demir, E., Sander, C. (2015).
    “PaxtoolsR: Pathway Analysis in R Using Pathway Commons.” 
    Bioinformatics.</li>
    <li>Sales G, Calura E, Cavalieri D, Romualdi C (2012). “graphite - a
    Bioconductor package to convert pathway topology to gene network.” BMC
    Bioinformatics. https://bmcbioinformatics.biomedcentral.com/articles/10 .
    1186/1471-2105-13-20.</li>
    <li>Guangchuang Yu, Li-Gen Wang, Yanyan Han and Qing-Yu He.clusterProfiler:
    an R package for comparing biological themes among gene clusters. OMICS:
    A Journal of Integrative Biology 2012, 16(5):284-287</li>
</ul>

# Session Info
```{r}
sessionInfo()
```