---
title: "HPO_vignette"
author:
- name: Erqiang Hu
  email: 13766876214@163.com
  affiliation: Department of Bioinformatics, School of Basic Medical Sciences, 
    Southern Medical University
date: "`r Sys.Date()`"
output:
  rmarkdown::html_vignette: default
  BiocStyle::html_document:
    toc: yes
    toc_float: yes
vignette: >
  %\VignetteIndexEntry{HPO_vignette}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

```{r setup}
library(HPO.db)
```




## Introduction
Human Phenotype Ontology (HPO) was developed to create a consistent description 
of gene products with disease perspectives,
and is essential for supporting functional 
genomics in disease context. Accurate disease descriptions can discover new 
relationships between genes and disease, and new functions for previous 
uncharacteried genes and alleles.We have developed the 
[DOSE](https://bioconductor.org/packages/DOSE/) package for semantic 
similarity analysis and disease enrichment analysis, and `DOSE` import an 
Bioconductor package 'DO.db' to get the relationship(such as parent and child)
between DO terms. But `DO.db`  hasn't been updated for years, and a lot of 
semantic information is [missing](https://github.com/YuLab-SMU/DOSE/issues/57). 
So we developed the new package `HPO.db` for Human Human Phenotype Ontology 
annotation.

##  :hammer: Installation

The released version from `Bioconductor`

```{r eval=FALSE}
if (!requireNamespace("BiocManager", quietly=TRUE))
    install.packages("BiocManager")
## BiocManager::install("BiocUpgrade") ## you may need this
BiocManager::install("HPO.db")
```

## Overview

```{r}
library(AnnotationDbi)
```

The annotation data comes from 
https://github.com/DiseaseOntology/HumanDiseaseOntology/tree/main/src/ontology,
and HPO.db provide these AnnDbBimap object:
```{r}
ls("package:HPO.db")
packageVersion("HPO.db")
```

You can use `help` function to get their documents: `help(HPOOFFSPRING)`
```{r}
toTable(HPOmetadata)
HPOMAPCOUNTS
```


## Fetch whole HPO terms

In HPO.db, `HPOTERM` represet the whole HPO terms and their names. The users can 
also get their aliases and synonyms from `HPOALIAS` and `HPOSYNONYM`, 
respectively.

convert HPOTERM to table
```{r}
doterm <- toTable(HPOTERM)
head(doterm)
```


convert HPOTERM to list
```{r}
dotermlist <- as.list(HPOTERM)
head(dotermlist)
```

get alias of `HP:0000006`
```{r}
doalias <- as.list(HPOALIAS)
doalias[['HP:0000006']]
```

get synonym of `HP:0000006`
```{r}
dosynonym <- as.list(HPOSYNONYM)
dosynonym[['HP:0000006']]
```


## Fetch the relationship between HPO terms
Similar to `HPO.db`, we provide four Bimap objects to represent relationship
between HPO terms: HPOANCESTOR,HPOPARENTS,HPOOFFSPRING, and HPOCHILDREN.

### HPOANCESTOR
HPOANCESTOR describes the association between HPO terms and their ancestral 
terms based on a directed acyclic graph (DAG) defined by the Human Phenotype
Ontology.
We can use `toTable` function in `AnnotationDbi` package to get a two-column 
data.frame: the first column means the HPO term ids, and the second column means 
their ancestor terms.
```{r}
anc_table <- toTable(HPOANCESTOR)
head(anc_table)
```

get ancestor of "HP:0000006"
```{r}
anc_list <- AnnotationDbi::as.list(HPOANCESTOR)
anc_list[["HP:0000006"]]
```


### HPOPARENTS
HPOPARENTS describes the association between HPO terms and their direct parent 
terms based on DAG. We can use `toTable` function in `AnnotationDbi` package to 
get a two-column data.frame: the first column means the HPO term ids, and the 
second column means their parent terms.
```{r}
parent_table <- toTable(HPOPARENTS)
head(parent_table)
```


get parent term of "HP:0000006"
```{r}
parent_list <- AnnotationDbi::as.list(HPOPARENTS)
parent_list[["HP:0000006"]]
```

###  HPOOFFSPRING
HPOPARENTS describes the association between HPO terms and their offspring  
terms based on DAG. it's the exact opposite of `HPOANCESTOR`, 
whose usage is similar to it.

get offspring of "HP:0000002"
```{r}
off_list <- AnnotationDbi::as.list(HPO.db::HPOOFFSPRING)
off_list[["HP:0000002"]]
```

### HPOCHILDREN
HPOCHILDREN describes the association between HPO terms and their direct 
children terms based on DAG. it's the exact opposite of `HPOPARENTS`, 
whose usage is similar to it.

get children of "HP:0000002"
```{r}
child_list <- AnnotationDbi::as.list(HPO.db::HPOCHILDREN)
child_list[["HP:0000002"]]
```

The HPO.db support the `select()`, `keys()`, `keytypes()`,
and `columns` interface.
```{r}
columns(HPO.db)
## use hpoid keys
dokeys <- head(keys(HPO.db))
res <- select(x = HPO.db, keys = dokeys, keytype = "hpoid", 
    columns = c("offspring", "term", "parent"))
head(res)
## use term keys
dokeys <- head(keys(HPO.db, keytype = "term"))
res <- select(x = HPO.db, keys = dokeys, keytype = "term", 
    columns = c("offspring", "hpoid", "parent"))   
head(res)
```


## Semantic similarity analysis
Please go to 
<https://yulab-smu.top/biomedical-knowledge-mining-book/> for the vignette.


## Disease enrichment analysis
Please go to 
<https://yulab-smu.top/biomedical-knowledge-mining-book/dose-enrichment.html> 
for the vignette.

```{r}
sessionInfo()
```