In oboforty/metafetcher: An R package for linking and standardizing metabolites data

{width=30% }

---

MetaFetcheR is an R package designed to link metabolites IDs from different Metabolome databases with eachother in a step to resolve ambiguity and standardize metabolites representation and annotation. Currently the package supports resolving IDs for the following databases:

• Human Metabolome Database (HMDB)
• Chemical Entities of Biological Interest (ChEBI)
• PubChem
• Kyoto Encyclopedia of Genes and Genomes (KEGG)
• Lipidomics Gateway (LipidMaps)

---

Installation

1. Install postgressql database on your system and create a user,you can download the database from here
2. Install devtools in R

install.packages("devtools")

1. Install MetafetcheR package

library(devtools)
devtools::load_all()
install_github("komorowskilab/metafetcher")

1. Download the database SQL dump files
2. HMDB
3. ChEBI

4. PubChem

5. Uncompress all downloaded files in a directory you create and name it tmp

6. Create a new R project and install MetafetcheR package

library(devtools)
devtools::load_all()
install_github("komorowskilab/metafetcher")
library(metafetcher)

1. call write_config with the

write_config(host,port,db_name,user,password,path_of_tmp_folder)

• host:"localhost" (this is the local host when you install postgres SQL)
• port: 5432 (this is the local port when you install postgres SQL)
• user: "postgres" (this is the local user that is created when you install postgresSQL)
• password: write the password that you want the data base to be created with
• path_of_tmp_folder: path to tmp folder that contains the extracted downloaded files.

8.call function install_database() for creating the tables and inserting the data from the tmp folder preferably put it in your R project directory

install_database() 

---

Example

Create a csv file with input IDs in the following format

library(knitr)
discovery=read.csv("discovery.csv")
kable(discovery,caption="example input table")

library(metafetcher)
df.res <- read.csv("discovery.csv", stringsAsFactors=FALSE)
resp <- resolve_metabolites(df.res)
print(resp$df)

df.out <- resp$df
kable(df.out ,caption="output table")

To map only a single ID you can use function resolve_single_id

library(metafetcher)
resp1 <- resolve_single_id('hmdb_id', 'HMDB0001005')

df.out1 <- resp1$df
print(df.out1)

kable(df.out1 ,caption="output table")

---

Citation

https://github.com/komorowskilab/metafetcher

---

oboforty/metafetcher documentation built on Aug. 18, 2021, 6:38 a.m.