ngreen-gt / lmm-formula-calculator Goto Github PK

This may cause some problems integrating the program with input data files, which can be in various mass formats (i.e. exact or ion mass). For now, this isn't being fixed, and exact masses should be used.

Hi,
the F,Cl,Br,I halogens are missing, hence outside use (medical, biomdecial, environmental,metabolomics) use is somewhat limited. Well silicon would be nice
because we already have Na.
Cheers
Tobias

Hi,
the complete formula should be generated as column according to the Hill notation,
so additional parsing can be avoided.
Cheers
Tobias

Hi,
tried to input Angiotensin C50H72N13O15P
Composition C:38-95 H:0-210 N:0-32 O:0-64 P:0-12 S:0-12
The input file input-angiotensin.dat

1124.5016
1125.5016
1126.5016

Angiotensin-Example

Z:\ChoFIT>chofit3_min.exe input-angiotensin.dat  output-angiotensin 100 1400 32 32 32 0


Program: CHOFIT3_MIN 20150315 Test Version
InFile : Z:\ChoFIT\input-angiotensin.dat
OutFile: Z:\ChoFIT\output-angiotensin.fit
Low MW :    100  High MW :   1400
Low C  :      1  High C  :      1
Low H  :      2  High H  :      2
Low O  :      0  High O  :      0
Low N  :      0  High N  :     24
Low S  :      0  High S  :      8
Low P  :      0  High P  :      4
Low Na :      0  High Na :      0
Low 13C:      0  High 13C:      0
Processing ID      2
Writing the output file...
Press ENTER to close the program...

Now the "high MW" numbers seem incorrect, but I am not sure how to set them.
Also the ouput file did not contain any formula that contained the correct solution of
C50H72N13O15P.

Where can I set the mass accuracy? Maybe that will solve it.
Maybe some more examples would help too, where do I set the limits for CHO etc.
Thanks
Tobias

Hi,
is it possible to generate tab separated output, in that way one can drag-and-drop the file directly into EXCEL without using regex replacement or EXCEL input filters.
Cheers
Tobias

The LMM code complicates finding multiple CHO formulae because it is optimized to stop after finding one. This normally works, but can fail if the mass tolerance is greater than 1.1 mDa, the minimum difference between two CHO formulae. The fix for this is a hybrid program which operates as is under 1.1 mDa tolerance, and search for the second CHO formulae when the tolerance is set higher. Because ultrahigh resolution data is typically below that mass tolerance, it doesn't affect it, but to work with lower resolution data it would be useful to make this fix.