Forwarding bug report from inchi-discuss: (1) handling of [H+], and (2) any way to short circuit InChIKey generation if given InChI?
- Noel ---------- Forwarded message ---------- From: Ken Smith <k.w.sm...@qub.ac.uk> Date: 7 September 2011 16:15 Subject: Re: [InChI-discuss] InChI for Proton and protonated hydrogen molecule To: inchi-disc...@lists.sourceforge.net Noel, Dmitrii, Geoff Many thanks for the very helpful info! Cheers, Ken P.S. Noel: OpenBabel converts [H+] to InChI=1S/H/q+1. According to Dmitrii (see below), I should be getting InChI=1S/p+1. On 7 Sep 2011, at 15:57, Noel O'Boyle wrote: > Speaking as an Open Babel dev, we should certainly look into > short-circuiting the generation of InChIKeys from InChIs. We do > generate the correct InChIKey though if given the input structure: > >> obabel -:"[H]1[H+][H]1" -oinchi -xK > ============================== > *** Open Babel Warning in InChI code > #1 :Accepted unusual valence(s): H(2) > RQZCXKHVAUFVMF-UHFFFAOYSA-N > 1 molecule converted > > - Noel > > On 7 September 2011 15:46, Dmitrii Tchekhovskoi > <dmitrii.tchekhovs...@nist.gov> wrote: >> Ken, >> >> This error message, "Problems/mismatches: Mobile-H( Charge(s): Do not >> match)" occurs when converting InChI=1S/H3/c1-2-3-1/q+1 to a chemical >> structure. >> >> I have no idea why Openbabel does this to produce an InChIKey from >> InChI. My >> guess is, makes the following conversions: >> InChI -> structure -> InChIKey >> >> The problem with this approach is that InChI cannot do the 1st step >> reliably >> for a small fraction of small molecule chemical space. This fact is >> mentioned in the InChI documentation. >> >> In your case, the failure occurred because the valence model >> implemented in >> the current version of InChI -> structure conversion does not >> number of >> bonds to a hydrogen atom greater than 1. >> >> If there is a way to make Openbabel convert a structure to InChI >> *AND* >> InChIKey at once, then try it. This approach is more reliable than >> the two >> step approach (first get InChI, secondly, get InChIKey from InChI >> via the >> reconstructed structure), which you use. Otherwise, you may use the >> software >> from the IUPAC web site. InChI software version 1.03 produced the >> following >> output for the triangular H3(+): >> >> InChI=1S/H3/c1-2-3-1/q+1 >> InChIKey=RQZCXKHVAUFVMF-UHFFFAOYSA-N >> >> Regards >> >> Dmitrii >> >>> -----Original Message----- >>> From: Ken Smith [mailto:k.w.sm...@qub.ac.uk] >>> Sent: Wednesday, September 07, 2011 10:07 AM >>> To: Tchekhovskoi, Dmitrii V. Dr. >>> Cc: inchi-disc...@lists.sourceforge.net >>> Subject: Re: [InChI-discuss] InChI for Proton and protonated >>> hydrogen >>> molecule >>> >>> Many thanks Dmitrii >>> >>> Assuming I go with the triangular structure, do you know how I can >>> generate the InChIKey? >>> >>> When using Openbabel to convert InChI to InChIKey, I get the >>> following >>> InChI software error when entering the triangular structure for H3+: >>> >>> InChI=1S/H3/c1-2-3-1/q+1 >>> >>> Problems/mismatches: Mobile-H( Charge(s): Do not match) >>> >>> and no key is generated. >>> >>> Cheers, >>> >>> Ken >>> >>> >>> >>> On 6 Sep 2011, at 20:43, Dmitrii Tchekhovskoi wrote: >>> >>>> In general, the correct InChI is the InChI produced by the software >>>> downloaded from the official IUPAC web site assuming one knows how >>>> to use >>>> it. >>>> If a structure has more than one atom then in order to distinguish >>>> isomers, >>>> InChI depends on the way the atoms in the structure are connected >>>> by >>>> chemical bonds. >>>> >>>>> -----Original Message----- >>>>> From: Ken Smith [mailto:k.w.sm...@qub.ac.uk] >>>>> Sent: Tuesday, September 06, 2011 9:36 AM >>>>> To: inchi-disc...@lists.sourceforge.net >>>>> Subject: [InChI-discuss] InChI for Proton and protonated hydrogen >>>>> molecule >>>>> >>>>> Folks >>>>> >>>>> Different tools I'm using to generate InChIs and InChIKeys are >>>>> presenting me a problem with protonated species. >>>>> >>>>> For example, which is the correct InChI for a proton? >>>>> >>>>> InChI=1S/H/q+1 >>>>> (ASSFXGJQJOXDAB-UHFFFAOYSA-N) >>>> >>>> [DT>>] -- Incorrect. This is just the Fixed H layer, which alone >>>> cannot be a >>>> standard InChI. >>>> *Please name the software you used to obtain this InChI and the >>>> options >>>> used.* This may be useful for other InChI users. >>>> >>>>> >>>>> or >>>>> >>>>> InChI=1S/p+1 >>>>> (GPRLSGONYQIRFK-UHFFFAOYSA-N) >>>> >>>> [DT>>] -- Correct >>>> >>>>> >>>>> >>>>> Likewise for protonated molecular hydrogen - which is correct. >>>>> Most >>>>> tools give me the first result below. >>>> >>>> [DT>>] -- Here InChI depends on the structure drawing: linear, >>>> triangular, >>>> or disconnected. >>>> >>>>> >>>>> InChI=1S/H3/h1H2/q+1 >>>> >>>> [DT>>] -- correct assuming linear connected structure H-H-H(+) or >>>> H- >>>> H(+)-H >>>> >>>>> >>>>> or >>>>> >>>>> InChI=1S/H2/h1H/p+1 >>>> >>>> [DT>>] Correct assuming H2 + H(+) disconnected structure. >>>> >>>> For a triangular structure, >>>> InChI=1S/H3/c1-2-3-1/q+1 >>>> >>>> >>>>> >>>>> >>>>> It's crucial that I get the latter one in particular correct, >>>>> since >>>>> it >>>>> plays a major role in our database. >>>>> >>>>> >>>>> Thanks in advance, >>>>> >>>>> Ken >>>>> >>>>> >>>>> >>>> >> ------------------------------------------------------------------------- >>> -- >>>>> --- >>>>> Special Offer -- Download ArcSight Logger for FREE! >>>>> Finally, a world-class log management solution at an even better >>>>> price-free! And you'll get a free "Love Thy Logs" t-shirt when you >>>>> download Logger. Secure your free ArcSight Logger TODAY! >>>>> http://p.sf.net/sfu/arcsisghtdev2dev >>>>> _______________________________________________ >>>>> InChI-discuss mailing list >>>>> inchi-disc...@lists.sourceforge.net >>>>> https://lists.sourceforge.net/lists/listinfo/inchi-discuss >>>> >>>> >> >> >> >> ------------------------------------------------------------------------------ >> Using storage to extend the benefits of virtualization and iSCSI >> Virtualization increases hardware utilization and delivers a new >> level of >> agility. Learn what those decisions are and how to modernize your >> storage >> and backup environments for virtualization. >> http://www.accelacomm.com/jaw/sfnl/114/51434361/ >> _______________________________________________ >> InChI-discuss mailing list >> inchi-disc...@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/inchi-discuss >> ------------------------------------------------------------------------------ Using storage to extend the benefits of virtualization and iSCSI Virtualization increases hardware utilization and delivers a new level of agility. Learn what those decisions are and how to modernize your storage and backup environments for virtualization. http://www.accelacomm.com/jaw/sfnl/114/51434361/ _______________________________________________ InChI-discuss mailing list inchi-disc...@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inchi-discuss ------------------------------------------------------------------------------ Using storage to extend the benefits of virtualization and iSCSI Virtualization increases hardware utilization and delivers a new level of agility. Learn what those decisions are and how to modernize your storage and backup environments for virtualization. http://www.accelacomm.com/jaw/sfnl/114/51434361/ _______________________________________________ OpenBabel-Devel mailing list OpenBabel-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-devel
