-
Notifications
You must be signed in to change notification settings - Fork 6
Expand file tree
/
Copy pathindex.Rmd
More file actions
46 lines (36 loc) · 2.08 KB
/
index.Rmd
File metadata and controls
46 lines (36 loc) · 2.08 KB
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
---
title: "SIDER 2"
author: "Daniel Himmelstein, Leo Brueggeman, Alexander Pico, Sergio Baranzini"
date: "February 6, 2015"
output:
html_document:
theme: cosmo
highlight: pygments
includes:
in_header: include/in_header.html
before_body: include/before_body.html
after_body: include/after_body.html
bibliography: references.bib
csl: cite-styles/plos-mod.csl
---
```{r, include=FALSE}
# Set up citation management
library(knitcitations)
knitcitations::cleanbib()
knitcitations::cite_options(citation_format = 'pandoc', cite.style = 'authoryear', hyperlink = 'to.bib', super = TRUE)
```
This website documents parsing [SIDER 2](http://sideeffects.embl.de) `r knitcitations::citep('10.1038/msb.2009.98')`, which is a database of indications and side effects extracted from drug labels. Using a similar method to an existing analysis `r knitcitations::citep('10.1126/science.1158140')`, we compute similarity measures between compounds.
## Components:
+ [**Parse**](parse) in *R* -- reading, manipulating, and performing quality control on the raw SIDER 2 data
+ [**Similarity**](similarity) in *R* -- calculating pairwise similarities between compounds based on their side effect and indication profiles
+ [**Compounds**](compounds) in *python* -- querying PubChem for additional compound information
## Downloads:
The [project repository](https://github.com/dhimmel/SIDER2) contains all files and code used for this project.
Several tab delimited text files are provided here for download:
+ [All side effects and indications, after processing](data/sider2-processed.txt). Each row represents an associated compound-concept pair.
+ [Pairwise compound similarities](data/similarities.txt.gz) for the three different similarity methods (based on indications, side effects, and substructure). The subset of compound-pairs that had values for all three methods is [also available](data/similarities-complete.txt.gz).
+ [Compound information](data/compounds.txt) including canonical smiles and parents.
# References
```{r, include=FALSE}
knitcitations::write.bibtex(file = 'references.bib')
```