This python script is one of the most convenient.

Change bigsdffilenameto your sdf file name (without .sdf)

and at the place of 100 write the number of molecules which you want in smaller file.

Don't forget to remove (number of molecules in each file--delete this line in bracket) from this file.

Try to convert the sdf file into mol2 or any other format which will store the compound individually rather than in a compiled one sdf file. Then, convert the generated mol2 files into many smaller sdf files for docking.

You could use the linux split function to split the file in as many subfiles you want. You may need to write a script (e.g. in Perl to make sure the split function splits your big file always at the end of a molecule description rather than randomly.

Dear Abdulrahman, Thank you for your reply, I had tried Babel software but it hang my system when I input the file. Your suggestion is most welcome if you know other tools and procedure.

Hi, This may give you some idea... it's the csplit function

http://autodock.scripps.edu/faqs-help/faq/is-there-a-way-to-save-a-protein-ligand-complex-as-a-pdb-file-in-autodock

This should help:

http://www.mayachemtools.org/docs/scripts/html/SplitSDFiles.html

It has a perl script along with all the options you can use to generate smaller sdf files containing n number of molecules in each file.

you can download the mayachemtools here:

http://www.mayachemtools.org/Download.html

I have checked out and it's working absolutely fine.

It was the second link that I got after googling with the words splitting sdf files

Try google first and if you don't find solution there then post your issue here.

this should help:

http://www.cgl.ucsf.edu/pipermail/chimera-users/2010-March/004974.html

split -a 1 -b 200m 2ge9_Model3_fin_2.txt

or you can use: Total Commander

Thank you NIcolas and Ravi Kanth. See I have no idea or very little idea about perl. I am from biological background. I am not pure bioinformatics guy. I have very limited time for accomplish my project. I cann't jump directly in computer oriented task. Our project is multi tasking and we have team. If any body who had passed through this type of task (like learning computer language from zero knowledge) and make me sure that this task will not take more then limited time, Then we go for it. More tech hardy task we put in another category and take help of other people. Google is best guru for us. Here problem is we are looking for GUI for first then easy programming and then Hard programming. Once again we have very limited time of aprox. three month(1 remaining). We have to report and justify significance work and write thesis and have to give viva.(we cant say we have no technical knowledge and we have just learn splitting of sdf that take month for learning perl!). My role is more about RDBMS, SQL and webinterface which I have completed almost so I am helping my colleagues for their issue. Research Gate people and Google help me lot.

You need not do much to run the script that i have mentioned. You just need to install perl, download the package given in the site and then run a one line command. Install perl in your system, download and extract the files into a folder and then let me know exactly how you want to break your file into smaller ones (i.e. how many entries per file) Then I will give you the command you need to run and how to run. It will be done with our much effort. Here are the links to download perl and the package from:

http://www.perl.org/get.html

http://www.mayachemtools.org/download/mayachemtools.zip

Let me know when you are done with installing perl and unzipping the package on to your system.

Thank you Abdulrahman and Vsyl for your valuable suggestion.

Again thank you Ravi Kanth for always being helpful to new bird. What we are aproching is we have 11 lakhs compunds and we are using molegrow to dock against 1 receptor. We have total 12 computer so we have to devide it in aprox ~1 lakh entry. I have installed biolinux 7 which have many inbound languages. I will get back to you .

Conceptually speaking, you may not want to divide the file using memory-blocks as a criterion, for you might break apart the first and last entries of each file. Perhaps it would be better to count the number of entries you have in the 3 gb file (you can use the "grep -c" command for this), and then break them up into 15 files, which should average around 200 mb. You could certainly write a script to break the file apart, but you could also use the "head" and "tail" commands that are built into Linux to accomplish this. It may take a little more time, but it is a much shorter learning curve. Of course, this all presupposes that each segment of data spans the same number of lines.

Thank you Joseph Norman! you got exact point I have to divide file in such way that end of file should contain whole-full molecule contained by it.

If you are accesible to Maestro, you can use following command in Linux system:

$SCHRODINGER/utilities/sdsubset -n range full.sdf > subset.sdf

Here, range specifies the first and last structure to extract, separated by a colon, e.g. 1:50000.

https://www.google.com.hk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=15&ved=0CEEQFjAEOAo&url=https%3a%2f%2fwww.schrodinger.com%2fDownload.php%3ftype%3dsupportdocs%26ident%3d528&ei=DOp2VKKeD9jf8AX2joF4&usg=AFQjCNGbM03uOY45xTANwIW6kiSwfQRSmw

Thank you Meng Fanwang...Thank you for detailed information.....

This python script is one of the most convenient.

Change bigsdffilenameto your sdf file name (without .sdf)

and at the place of 100 write the number of molecules which you want in smaller file.

Don't forget to remove (number of molecules in each file--delete this line in bracket) from this file.

How do I split an SD file of 35 separate structures into 35 individual structure files to use within a docking program (PyRx). PDB, PDBQT or MOL2 would probably be best. I've seen that this is possible with openBabel, but no good explanations on how to do it using the openBabel GUI interface. I'm not good enough with command line operations to figure out how to do it using babel with command prompt. Thank you.