{ "version": "https://jsonfeed.org/version/1.1", "title": "chem-bla-ics", "description": "Chemblaics (pronounced chem-bla-ics) is the science that uses open science and computers to solve problems in chemistry, biochemistry and related fields.", "home_page_url": "https://chem-bla-ics.linkedchemistry.info/", "feed_url": "https://chem-bla-ics.linkedchemistry.info/2008/05/03/wicked-chemistry-and-unit-testing.json", "icon": "https://chem-bla-ics.linkedchemistry.info/assets/images/chem-bla-ics_logo.png", "language": "en", "authors": [ { "name": "Egon Willighagen", "url": "https://orcid.org/0000-0001-7542-0286", "_orcid": "0000-0001-7542-0286" } ], "items": [ { "id": "https://doi.org/10.59350/fwsac-99191", "url": "https://chem-bla-ics.linkedchemistry.info/2008/05/03/wicked-chemistry-and-unit-testing.html", "title": "Wicked chemistry and unit testing", "content_html": "

After a discussion on starting development releases for CDK on cdk-devel,\nthe discussion continued on the state of the CDK atom typer .\nDan and Rajarshi\nhave done tests in the past against PubChem and its DTP/NCI subset. Rajarshi made his\nanalysis part of CDK Nightly,\nand provides but a summary\n(which seems broken: zero fails) and a detailed list.

\n\n

Dan, do I understand correctly that those Structure Evaluation:No Comparision - Unparameterized Atom - S. lines in the\nDepositor-Supplied Comments section on PubChem are based on CDK trunk? That would be a great honor! Anyways…

\n\n

The amount of atom types we use to describe the chemistry we observe is overwhelming (even without charged or radical atoms).\nAnd, most atom type lists are quite limited in what they represent. However, having an explicit list allows the computer to\ndecide if it can do reasonable calculations on a structure. Always filter your data to screen for unrecognized atom types,\nbefore heading of to, for example, QSAR calculations!

\n\n

Now, many fails are because of the incomplete CDK atom type list (e.g. Au in SID:413374),\nor because the atom typer code has a bug (e.g. SID:403517).\nAnd these screenings against PubChem provide a nice priority list. However, others are either because the used SDF format\ncannot represent the chemistry (e.g. SID:420394), or the\nentry is a plain wrong (e.g. SID:301178). The latter\ntwo types of fails, I am annotating using del.icio.us/egonw/pubchem+check-valency\nfor others to comment on (just tag the same page using del.ici.us, and I’ll see the comments show up.

\n\n

Unit testing

\n\n

For the first two types of fails, basically three things need to be done:

\n\n\n\n

Because we cannot use SMILES or file readers for writing these tests (than we can confounding\nof error sources), we have to hard code the chemical structure, which may be a bit cumbersome.

\n\n

Unless you use the CDKSourceCodeWriter!\nThis IChemObjectWriter\ncreates CDK source code, staring with a IMolecule. Now, because our bug reports are derived from fails against the PubChem\nscreening, we can simply use this BeanShell code to download a structure from\nPubChem and convert it to CDK source code:

\n\n
#!/usr/bin/bsh\n\nimport org.openscience.cdk.Molecule;\nimport org.openscience.cdk.io.MDLV2000Reader;\nimport org.openscience.cdk.io.CDKSourceCodeWriter;\n\nif (bsh.args.length == 0 || bsh.args[0] == null) {\n  System.out.println(\"Syntax: pubchem2unittest.bsh [CID]\\n\");\n  System.exit(0);\n}\n\nString cid = bsh.args[0];\nString urlString = \"http://pubchem.ncbi.nlm.nih.gov/summary/summary.cgi?disopt=SaveSDF&cid=\" + cid;\n\nURL url = new URL(urlString);\n\nMDLV2000Reader reader = new MDLV2000Reader(url.openStream());\nMolecule mol = reader.read(new Molecule());\n\nStringWriter stringWriter = new StringWriter();\nCDKSourceCodeWriter writer = new CDKSourceCodeWriter(stringWriter);\nwriter.write(mol);\nwriter.close();\n\nSystem.out.print(stringWriter.toString());\n
\n\n

For example, I am currently debugging a sulphur atom type perception problem, for which the simplest\nsubstructure looks like (sid=12279910 ,\nInChI=1/C2H7NS/c1-4(2)3/h3H,1-2H3):

\n\n

\"\"

\n\n

I can convert this PubChem entry to CDK source code with:

\n\n
$ bsh -classpath dist/jar/cdk-svn-20080221.jar tools/pubchem2unittest.bsh 12279910\n
\n\n

Resulting in this output which I can copy/paste into my unit test:

\n\n
IMolecule mol = new Molecule();\nIAtom a1 = mol.getBuilder().newAtom(\"S\");\na1.setPoint2d(new Point2d(2.866, 0.25));  mol.addAtom(a1);\nIAtom a2 = mol.getBuilder().newAtom(\"N\");\na2.setPoint2d(new Point2d(3.7321, 0.75));  mol.addAtom(a2);\nIAtom a3 = mol.getBuilder().newAtom(\"C\");\na3.setPoint2d(new Point2d(2.0, 0.75));  mol.addAtom(a3);\nIAtom a4 = mol.getBuilder().newAtom(\"C\");\na4.setPoint2d(new Point2d(2.866, -0.75));  mol.addAtom(a4);\nIAtom a5 = mol.getBuilder().newAtom(\"H\");\na5.setPoint2d(new Point2d(2.31, 1.2869));  mol.addAtom(a5);\nIAtom a6 = mol.getBuilder().newAtom(\"H\");\na6.setPoint2d(new Point2d(1.4631, 1.06));  mol.addAtom(a6);\nIAtom a7 = mol.getBuilder().newAtom(\"H\");\na7.setPoint2d(new Point2d(1.69, 0.2131));  mol.addAtom(a7);\nIAtom a8 = mol.getBuilder().newAtom(\"H\");\na8.setPoint2d(new Point2d(2.246, -0.75));  mol.addAtom(a8);\nIAtom a9 = mol.getBuilder().newAtom(\"H\");\na9.setPoint2d(new Point2d(2.866, -1.37));  mol.addAtom(a9);\nIAtom a10 = mol.getBuilder().newAtom(\"H\");\na10.setPoint2d(new Point2d(3.486, -0.75));  mol.addAtom(a10);\nIAtom a11 = mol.getBuilder().newAtom(\"H\");\na11.setPoint2d(new Point2d(4.269, 0.44));  mol.addAtom(a11);\nIBond b1 = mol.getBuilder().newBond(a1, a2, DOUBLE);\nmol.addBond(b1);\nIBond b2 = mol.getBuilder().newBond(a1, a3, SINGLE);\nmol.addBond(b2);\nIBond b3 = mol.getBuilder().newBond(a1, a4, SINGLE);\nmol.addBond(b3);\nIBond b4 = mol.getBuilder().newBond(a2, a11, SINGLE);\nmol.addBond(b4);\nIBond b5 = mol.getBuilder().newBond(a3, a5, SINGLE);\nmol.addBond(b5);\nIBond b6 = mol.getBuilder().newBond(a3, a6, SINGLE);\nmol.addBond(b6);\nIBond b7 = mol.getBuilder().newBond(a3, a7, SINGLE);\nmol.addBond(b7);\nIBond b8 = mol.getBuilder().newBond(a4, a8, SINGLE);\nmol.addBond(b8);\nIBond b9 = mol.getBuilder().newBond(a4, a9, SINGLE);\nmol.addBond(b9);\nIBond b10 = mol.getBuilder().newBond(a4, a10, SINGLE);\nmol.addBond(b10);\n
", "summary": "After a discussion on starting development releases for CDK on cdk-devel, the discussion continued on the state of the CDK atom typer . Dan and Rajarshi have done tests in the past against PubChem and its DTP/NCI subset. Rajarshi made his analysis part of CDK Nightly, and provides but a summary (which seems broken: zero fails) and a detailed list.", "image": "https://chem-bla-ics.linkedchemistry.info/assets/images/sid12279910.png", "date_published": "2008-05-03T00:00:00+00:00", "date_modified": "2025-10-11T00:00:00+00:00", "tags": ["cdk","pubchem"], "authors": [ { "name": "Egon Willighagen", "url": "https://orcid.org/0000-0001-7542-0286" } ] } ] }