Lambda notation bug?

[dan2097]

Lambda notation bug?

[dan2097]

Original comment by Daniel Lowe (Bitbucket: dan2097, GitHub: dan2097).

---

I don't think it's a bug. The nitrogen in morpholine is in its standard valency so you correctly don't need to employ the lambda convention. But this means the name can be written simply as:
1lambda6-thiomorpholine-1,1-dione

The different from the 1lambda6,2-benzothiazole-1,1-dione case is that benzothiazole contains a Hantzsch-Widman ring which should have the position of its heteroatoms given e.g. 1lambda6,2-benzothiazole-1,1-dione is different from 1lambda6,3-benzothiazole-1,1-dione. thiomorpholine on the other hand is a trivial name and always has the sulfur at position 1/ nitrogen at position 4. (you could argue it's semi-systematic and is a modification of the trivial name morpholine)

I will report this bug to ChemAxon ;-)

[dan2097]

Original comment by pmortenson (Bitbucket: pmortenson, ).

---

Thanks, Dan, that sounds pretty clear cut. As you correctly surmised, we use ChemAxon to generate names, and then round trip them with OPSIN as a check. When the round trip doesn't work, I have to try to assign blame....

The 1,3 isomer is correctly named as a thiazinane, which OPSIN happily parses. So I'll kick this one over to Hungary!

All the best,
Paul

[dan2097]

Original comment by Daniel Bonniot de Ruisselet (Bitbucket: [Daniel Bonniot](https://bitbucket.org/Daniel Bonniot), ).

---

Paul, Daniel,

I can confirm this was a bug in ChemAxon Structure-to-Name. The second locant (4) was printed by mistake as an unanticipated consequence of printing the lambda locant.

I fixed this in our trunk, so this will be released in our 6.2 version. The generated name is now as expected: 1λ⁶-thiomorpholine-1,1-dione

In my tests this improves the generated name in roughly 0.2% of the compounds in a large diverse set, which is nothing to sneeze at :) Thanks for discovering and reporting this issue.

Daniel Bonniot, ChemAxon

[dan2097]

Original comment by Daniel Lowe (Bitbucket: dan2097, GitHub: dan2097).

---

Daniel has resolved this issue :)

[dan2097]

Original comment by pmortenson (Bitbucket: pmortenson, ).

---

Thanks to both for your efforts!