Open .mol files and compiling information - r

Question

I'm trying to create a program that opens a lot of files (.mol), and copies specific information from those files and saves it into a spreadsheet (TAB delimited files '\t').
I have 10000 mol files on my computer that look like SN00000001 SN00000002 SN00000003 ... SN00010000.
(download link => http://bioinf-applied.charite.de/supernatural_new/src/download_mol.php?sn_id=SN00000001)
I have two questions:
I already tried to use function load.molecules (rcdk) and ChemmineR (loadsdf) but I did not succeed to open a .mol file in R.
It´s possible to open each .mol file and save specific information such as "ID", "Name", "Molecular Formula" it into a unique spreadsheet using R?

Answer 1

Ok, I will send you the code
# get the full path of your mol files
mol_files <- list.files(path = file.path(getwd(), "/Users/189919604/Desktop/Download
SuperNatural II/SN00000001"), # specify your folder here
pattern = "*mol",
full.names = TRUE)
# create tibble, with filenames (incl. the full path)
df <- tibble(filenames = mol_files)
# create function to extract all the information
extract_info <- function(sdfset) {
# function to extract information from a sdfset (ChemmineR)
# this only works if there is one molecule in the sdfset
ID <- sdfset#SDF[[1]]#datablock["SNID"]
Name <- sdfset#SDF[[1]]#header["Molecule_Name"]
Molecular_Formula <- sdfset#SDF[[1]]#datablock["Molecular Formula"]
sdf_info <- tibble(SNID = ID,
Name = Name,
MolFormula = Molecular_Formula)
return(sdf_info)
}
# read all files and extract info
df <- df %>%
mutate(sdf_data = map(.x = filenames,
.f = ~ read.SDFset(sdfstr = .x)),
info = map(.x = sdf_data,
.f = ~ extract_info(sdfset = .x)))
# make a nice tibble with only the info you want
all_info <- df %>%
select(molecule) %>%
unnest(info)
# write to file
write_delim(x = all_info,
path = file.path(getwd(), "test.tsv"),
delim = "\t")

Answer 2

I hope this works, I only tested it with 2 mol files. I used read.SDFset from ChemmineR package to read all the mol files. The package tidyverse I use, is to work with tibbles. Tibbles are actually dataframes with some extra properties / functionalities.
library(tidyverse)
library(ChemmineR)
# get the full path of your mol files
mol_files <- list.files(# specify your folder here in case of windows also add your drive letter e.g.: "c:/users/path/to/my/mol_files"
path = "/home/rico/r-stuff/temp",
pattern = "*mol",
full.names = TRUE)
# create tibble, with filenames (incl. the full path)
df <- tibble(filenames = mol_files)
# create function to extract all the information
extract_info <- function(sdfset) {
# function to extract information from a sdfset (ChemmineR)
# this only works if there is one molecule in the sdfset
ID <- sdfset#SDF[[1]]#datablock["SNID"]
Name <- sdfset#SDF[[1]]#header["Molecule_Name"]
Molecular_Formula <- sdfset#SDF[[1]]#datablock["Molecular Formula"]
sdf_info <- tibble(SNID = ID,
Name = Name,
MolFormula = Molecular_Formula)
return(sdf_info)
}
# read all files and extract info
df <- df %>%
mutate(sdf_data = map(.x = filenames,
.f = ~ read.SDFset(sdfstr = .x)),
info = map(.x = sdf_data,
.f = ~ extract_info(sdfset = .x)))
# make a nice tibble with only the info you want
all_info <- df %>%
select(info) %>%
unnest(info)
# write to file
write_delim(x = all_info,
path = file.path(getwd(), "temp", "test.tsv"),
delim = "\t")