problem with metals and memory

Hello,

I have started using metabolizer and I am interested in some batch mode. I've faced some problems with compounds having metals in the SMILES, such as:

[Al](CC(C)C)(CC(C)C)CC(C)C

[Co+2][OH+][C-]([O-])CCCCCCCCCCCCCCCCC

Those two work when used in interfaced version of metabolizer but when using command window I get info:

(CC(C)C)(CC(C)C)CC(C)C was unexpected at this time

or CCCCCCCCCCCCCCCCC was unexpected at this time

 

I would be grateful for some help here.

 

And one small thing that I have faced after dealing with few hundreds of compounds (and writing them out as separated files) is reaching java heap space for memory (java.lang.OutOfMemoryError: Java heap space). May be stupid question but how can I clead the memory? I don't want to expand memory given for Java (as far as I know I can do that).

 

Thank you very much!

I tried using both of your SMILES strings and they executed quickly with the default settings. This is how I called the command line tool:

metabolize '[Al](CC(C)C)(CC(C)C)CC(C)C'

How do you run Metabolizer? What are your command line options?

Please make sure to enclose the SMILES string in quotes if you have special characters in them (such as brackets), because the command line interpreter can misinterpret them.

You can run out of memory in case of enumerating multiple generations in exhaustive mode. The default java memory is usually small, and it is easy to create thousands of metabolites in few exhaustive steps depending on your input structure an settings. Metabolites are kept in memory during enumeration to avoid loops and to be able to detect duplicates considering multiple routes to the same metabolite (in global mode).

Furthermore, Metabolizer always calculates +1 generation of the metabolites to determine bothe the formation and degradation steps of the metaboites as well, which are needed for the calculation of the accumulation indicator. So, if you set the generation limit to 3 generations, a 4th generation is calculated as well in the background.

The memory consumption will be decreased in the near future, but the danger of combinatorial explosion still will be there in case of many generations. However, this risk will be decreased. Version 6.2 will be released today, however, if you can try it, does it change the situation?

Hello

Thank you for your answer. When I've quotes on both sides of SMILES metabolizer can read the SMILES with metals, however when uploading smiles I use smi files (with a list of smiles). Then uploading of chemicals with metals doesn't work (other chemicals work). I've tried using commas in smi file but it didn't make any difference. 

 

the command I use is:

metabolize in_file.smi -lg 3 -g -ga 0.1 -gn 1 -o out_file.txt -m true

 

When it comes to the memory peak I found that this was the problematic compound:

O1C(CO)C(OC2OC(CO)C(OC3OC(CO)C(OC4OC(CC(O)C4O)CO)C(O)C3O)C(O)C2O)C(O)C(O)C1OC1C(O)C(O)C(OC1CO)OC1C(O)C(O)C(OC1CO)OC1C(O)C(O)C(OC1CO)O

Metabolizer crashed when putting this chemical in both command and interface version. There were hundrets of chemicals predicted in 3rd generation (what I have noticed). I guess the solution is to use only 2 generations?

 

I'm happy to see that new version was released. I must say though that I got very strange effect of installing the new version. I've runned instalator normally and ask for putting new JChem in the same location as the previous one. I got question about uninstalling the old version - I've agreeded. Old version got uninstalled, new version got installed. Everything seem to be ok, I can open interface metabolizer without a problem but now command 'metabolize' doesn't work att all in the command window. When I write the command:

start /B metabolize

I get info:

Windows cannot find 'metabolize'. Make sure you typed the name correctly, and then try again.

and when writing commands:

'metabolize' is not recognized as an interval or external command, operable program or batch file.

I didn't change anything apart from uninstalling old version of JChem and installing new version in the same location. I didn't install new version of Marvin Beans. And I've restarted computer if that matters at all.

 

I would be grateful for some help here, thank you!

I just wanted to say that  that computer programs are sometimes strange and I don't understand why they need to be double uninstalled and installed in order to work. Nevertheless, I don't know if you did something but the new version of metaboliser deals with the metal problem that I've described. No problem now!