Menu
▾
▴
Re: [Psidev-pi-dev] Calculating amino acid average masses
|
From: David C. <dc...@ma...> - 2008-10-13 17:50:46
|
Hi Matt, Ah yes... looks as though we've not updated the help page since we made the change to use unimod. You can see the residue masses calculated in Unimod by searching for <umod:amino_acids> in the xml here: http://www.unimod.org/xml/unimod.xml F: 147.1739 Y: 163.1733 T: 101.1039 Which are the same as the NIST values (but rounded). Sorry about that, we'll update the help page.... David Matthew Chambers wrote: > Hi David, > > The masses on this page http://www.matrixscience.com/help/aa_help.html > do not appear to use the IUPAC atomic weights (which are the same as > the NIST atomic weights). It's hard to tell because of the rounding, > but several amino acids seem to be rounding off according to the > Waters avg or at least something not quite the NIST avg: > > Waters > NIST > Mascot > F 147.176600 147.173860 147.18 > Y > 163.176000 163.173260 163.18 > > > Then there's this residue, which strangely rounds with NIST, so > perhaps it is using the 12.01085 number Brendan mentioned. > T 101.105100 101.103880 101.10 > > > -Matt > > > David Creasy wrote: >> Hi Matt, >> >> Unimod (and hence Mascot) calculates residue masses from IUPAC atomic >> weights: >> >> http://www.unimod.org/masses.html >> http://www.webelements.com/periodicity/atomic_weight/ >> >> We are 'skirting' the problem in analysisXML by allowing the mass >> values used by the search engine to be recorded in the analysisXML >> instance documents. Any software that reads analysisXML documents >> should probably use those values. >> >> David >> >> Matthew Chambers wrote: >>> Hi all, >>> >>> Some discussion has come up within the ProteoWizard group and some >>> collaborators (Jimmy Eng and Brendan MacLean) about how to calculate >>> amino acid average masses. Currently, we are using the atomic >>> weights from NIST: >>> http://physics.nist.gov/cgi-bin/Compositions/stand_alone.pl?ele=&all=all&ascii=html&isotype=some >>> >>> >>> We calculate average mass (molecular weight) of a chemical formula >>> in ProteoWizard by <count of element> * <atomic weight of element>. >>> But when we calculate the average masses of the amino acids and then >>> compare these to the online references for amino acid masses, the >>> discrepancies are (IMO) unacceptable. >>> >>> A table showing some different values and references: >>> expasy = >>> http://education.expasy.org/student_projects/isotopident/htdocs/aa-list.html >>> >>> waters = a laminated reference sheet I have from Waters; it uses >>> these atomic weights: C=12.011 H=1.00794 O=15.9994 N=14.00674 S=32.066 >>> pwiz avg = element count * NIST atomic weight >>> >>> >>> >>> Waters mono Waters avg expasy avg pwiz mono >>> pwiz avg Waters - pwiz for mono Waters - pwiz for avg >>> Alanine A 71.037110 71.078800 71.078800 >>> 71.037114 71.077900 -0.000004 0.00090 >>> Arginine R 156.101110 156.187600 156.187500 >>> 156.101111 156.185680 -0.000001 0.00192 >>> Asparagine N 114.042930 114.103900 114.103800 >>> 114.042927 114.102640 0.000003 0.00126 >>> Asparticacid D 115.026940 115.088600 115.088600 >>> 115.026943 115.087400 -0.000003 0.00120 >>> Cysteine C 103.009190 103.144800 103.138800 >>> 103.009184 103.142900 0.000006 0.00190 >>> Glutamicacid E 129.042590 129.115500 129.115500 >>> 129.042593 129.113980 -0.000003 0.00152 >>> Glutamine Q 128.058580 128.130800 128.130700 >>> 128.058578 128.129220 0.000002 0.00158 >>> Glycine G 57.021460 57.052000 57.051900 >>> 57.021464 57.051320 -0.000004 0.00068 >>> Histidine H 137.058910 137.141200 137.141100 >>> 137.058912 137.139280 -0.000002 0.00192 >>> Isoleucine I 113.084060 113.159500 113.159400 >>> 113.084064 113.157640 -0.000004 0.00186 >>> Leucine L 113.084060 113.159500 113.159400 >>> 113.084064 113.157640 -0.000004 0.00186 >>> Lysine K 128.094960 128.174200 128.174100 >>> 128.094963 128.172280 -0.000003 0.00192 >>> Methionine M 131.040490 131.198600 131.192600 >>> 131.040485 131.196060 0.000005 0.00254 >>> Phenylalanine F 147.068410 147.176600 147.176600 >>> 147.068414 147.173860 -0.000004 0.00274 >>> Proline P 97.052760 97.116700 97.116700 >>> 97.052764 97.115180 -0.000004 0.00152 >>> Serine S 87.032030 87.078200 87.078200 >>> 87.032028 87.077300 0.000002 0.00090 >>> Threonine T 101.047680 101.105100 101.105100 >>> 101.047678 101.103880 0.000002 0.00122 >>> Tryptophan W 186.079310 186.213300 186.213200 >>> 186.079313 186.209900 -0.000003 0.00340 >>> Tyrosine Y 163.063330 163.176000 163.176000 >>> 163.063329 163.173260 0.000001 0.00274 >>> Valine V 99.068410 99.132600 99.132600 >>> 99.068414 99.131060 -0.000004 0.00154 >>> >>> >>> avg mono delta: -0.00000110 >>> avg avg delta: 0.001756 >>> >>> For a peptide of 20 residues, the average delta for average mass >>> calculations between the NIST atomic weights and the Waters atomic >>> weights would be: 0.03512 >>> Not a huge number in the average mass domain, but enough to make me >>> worry. >>> >>> I think it's pretty clear that the 12.011 value for carbon is closer >>> to the atomic weight used in the amino acid references that are >>> widely used in the field. Several other references which I couldn't >>> easily copy and paste also used similar average masses, e.g.: >>> http://www.matrixscience.com/help/aa_help.html >>> http://www.i-mass.com/guide/aamass.html (this looks like a scan of >>> the Waters reference, same format and values) >>> http://musclesurf.com/beverly-mass-amino.html (this is not an amino >>> acid mass reference, but included for completeness) ;) >>> >>> More importantly, I couldn't find any amino acid average mass >>> references that looked like they were calculated from the NIST >>> atomic weights. >>> >>> So the question is: what atomic weights should researchers use to >>> calculate amino acid masses? I suspect most software just stores the >>> amino acid masses directly and avoids the problem, but it's an >>> interesting issue to address for the PSI and more specifically the >>> PSI-PI. :) >>> >>> Thanks, >>> Matt >>> ------------------------------------------------------------------------ >>> >>> >>> ------------------------------------------------------------------------- >>> >>> This SF.Net email is sponsored by the Moblin Your Move Developer's >>> challenge >>> Build the coolest Linux based applications with Moblin SDK & win >>> great prizes >>> Grand prize is a trip for two to an Open Source event anywhere in >>> the world >>> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >>> ------------------------------------------------------------------------ >>> >>> >>> _______________________________________________ >>> Psidev-pi-dev mailing list >>> Psi...@li... >>> https://lists.sourceforge.net/lists/listinfo/psidev-pi-dev >>> >> >> -- >> David Creasy >> Matrix Science >> 64 Baker Street >> London W1U 7GB, UK >> Tel: +44 (0)20 7486 1050 >> Fax: +44 (0)20 7224 1344 >> >> dc...@ma... >> http://www.matrixscience.com >> >> Matrix Science Ltd. is registered in England and Wales >> Company number 3533898 -- David Creasy Matrix Science 64 Baker Street London W1U 7GB, UK Tel: +44 (0)20 7486 1050 Fax: +44 (0)20 7224 1344 dc...@ma... http://www.matrixscience.com Matrix Science Ltd. is registered in England and Wales Company number 3533898 |