Discussion | DivCHED CCCE: Cheminformatics OLCC

Question on Provenance

How far along is digital integration of lab notebooks? I would assume that this would help a lot considering we are now dealing with computers for nearly every aspect of the experiment process.

PubChem does not appear to display CAS numbers for some compounds, such as 1-aminoethane-1,2-diol. Is there a reason for this (such as private access) and how can one go about retrieving a CAS number?

Resolver and 1.3.3a

Hi John, Resolver does not really compute structures. It's a super fast approximation of reasonable geometry. This is done by first using an algorithm to break a connectivity identifier (SMILES, IUPAC, InChI) into library building blocks called ensembles. The ensembles are then reassembled into the approximate 3d model, carrying with them their respective geometries. One quick way to see the approximate nature of this is to measure bond angles in simple substituted alkenes. For example, load 2-methylprop-1-ene from PubChem and NIH/CADD Resolver and check the C-C-C bond angles. The "computed" 3D models that we get from databases all carry the weight of the their respective calculation algorithms. Take a look at the 3D model of allene fromPubChem and NIH/CADD Resolver. Otis

Re: Assignment Exercise 2

Hi Joshua, You've read it in the "correct" (i.e. the usual, conventional, expected) direction. Good! However, try drawing the structure based on reading the contracted formula from left to right, starting with the H3C methyl group and working your way toward the ring system, and see what you end up with. Evan

Re: Stereochemistry question b

John, Are you sure you're getting different structures? Or rather, are you getting 1) **different R/S configurations** of stereocenters, or 2) **different ways of drawing the same** R/S configuration? (You may be getting multiple tautomers, which is worth noting. But if the tautomerism doesn't involve stereocenters, then the "expected stereochemistry" would be a consistent R/S configuration for each of the tautomers.) Evan

Re: Stereochemistry

Hi John, The short answer: both structures display the same, *expected* stereochemistry. In terms of R/S configuration of stereocenters, the wedged ring bond (with H left implicit) is the same as a dashed bond to H (with the ring bond in the plane of the page). You can verify this for yourself by building a model. (I try to keep four distinct pens and pencils with me to use in making ad-hoc models of stereocenters....) The longer answer: you're right that these structures sure **look** different, even if they convey the same chemical information in terms of connectivity and stereocenter configuration. If you're communicating with other (human) chemists, you'll want to depict stereochemistry consistently - when you're dealing with rings, it's often easiest to keep the entire ring in the plane of the page and dash or wedge hydrogens and substituents, even though only a few rings (e.g. benzene, cyclopropane) actually have planar conformations. Evan

Module 4a: Assignment Exercise 2

Hi Dr's McEwen and Hepler-Smith, I'm confused about what you mean when you say "How does the full structural formula appear if the label is read the other way?" What is the other way the contracted formula can be read? I read it right to left.

Stereochemistry question b

For 1.3.3., question b, I am receiving different resolved structures from Chemspider and NIH. I am assuming that this indicates the expected stereochemistry as not being represented.

Stereochemistry

For 1.3.3. question a, the 2-D resolved structure from opsin indicates that the cyclized oxygen is coming out of the plane. On the other hand the 3-D resolved structure from NIH shows a planar ring. Would this qualify as the stereochemistry not showing up as expected?

You may find interesting to

You may find interesting to check the Chemistry Add-in for Word: <a href="https://chem4word.codeplex.com/">https://chem4word.codeplex.com/</a>. I haven't used it myself but it does look promising.