1. Kyle Finchsigmate on June 21, 2006 10:27 PM writes...

I find blogging to be a horrible waste of time.

2. Petros on June 22, 2006 1:50 AM writes...

I recollect a HTS that was looking for a subtype selective ligand by screening, via radioligand binding assays, against three commercially provided cloned receptors. Some 200,000 compounds later it turned out the vendor had mislabelled the subtype of interest and that the screening had been aaginst the wrong subtype. All due to molecular biologist's mistake apparently.

3. Peter Ellis on June 22, 2006 3:13 AM writes...

Well, there was a famous case in the UK at the height of the BSE scare. Some lab owned up to having spent X zillion pounds of the nation's money testing sheep brains for a cattle disease. "Bovine" and "Ovine" brain extract look so similar on the label...

4. Denni on June 22, 2006 4:33 AM writes...

Beware mycoplasma--especially in stock solutions!

5. Kay on June 22, 2006 7:44 AM writes...

Any assay that portends to predict human PK.

6. ss on June 22, 2006 7:53 AM writes...

what about the chemist who made the wrong regioisomer/enantiomer/diasteromer... and sent it for evalauation ?

7. Jim Hu on June 22, 2006 8:41 AM writes...

This is a story of how a waste of time had a silver lining. Way back in the stone age when I was a beginning grad student, it was before efficient DNA sequencing, and even before widely available restriction enzymes. There was a rumor - it might have even been published somewhere - that the Rous Sarcoma Virus provirus did not have any AvaI sites. This was important, as it offered a strategy for cloning the RSV genome. As it turns out, there are AvaI sites in RSV, so the efforts of large numbers of virology labs all over the world to purify AvaI from Anabaena (a cyanobacterium; think green pond scum) would fall into the waste of time category.

These labs, including one I was in at the time, were also not experts in large scale culture of pond scum. A postdoc in my lab wound up purifying a reasonably large amount of a restriction enzyme from a contaminating bacterium. This was not useful for cloning the RSV provirus either, and it wasn't even a good prep of AvaI. Another waste of time.

However, as this was back in the days of chemical sequencing short distances from restriction sites by the Maxam -Gilbert methods, the enzyme was duly stored. A friend in another lab was desperately trying to find a site that would allow him to close a gap in the sequence of the bacteriophage λ origin of replication. Guess what was the only available enzyme that cut in the needed interval!

8. Derek Lowe on June 22, 2006 9:11 AM writes...

ss, sometimes that "wrong" isomer turns out to be more active than the one that was being targeted, y'know. But when it comes to assays, chemists waste time on the retail level only. For wholesale, you have to go to the other departments. (Which is not to say that we don't have our own bulk supplies of futility in other places. . .)

9. Hap on June 22, 2006 10:56 AM writes...

It's not a horrible waste of time, but my first reaction in grad school was protecting 1,6-hexanediamine with Boc. I used BOC-ON for the protection and isolated...ON. Oops.

I also managed to waste way too much time trying to monoprotect 1,5-cyclooctanediol with benzyl bromide and spending time on fruitless reactions to do it. One person suggested KOH in DMSO which gave a nice yellow color but didn't help because our high-vac couldn't pull off DMSO. Out of all of this all I managed to do was to sensitize myself to benzyl bromide.

10. qetzal on June 22, 2006 11:25 AM writes...

I don't have a story about assays per se, but I have one that's in a similar vein.

About 20 years ago, I heard there was a lab that spent enormous effort trying to determine the crystal structure of some interesting bacterial protein. They spent a long time purifying it, coaxing it to form high quality crystals, getting good X-ray diffraction patterns, and deriving the structure.

According to the story, they were presenting their results at a meeting when someone in the audience noticed it looked a lot like the crystal structure for lysozyme.

Turns out, they had used lysozyme to crack open the bacteria, and had inadvertantly purified and crystallized the lysozyme instead of the protein of interest.

Don't know if it's really true, but I always thought it was a good story.

11. Mark on June 22, 2006 12:25 PM writes...

RE: number 6 above,

What about the FTE at the CRO who makes the wrong isomer and is too dull to correctly interpret the NMR, scales it up and carries it fwd over 3 months and 8 steps later only to have it fail horrible in a key cyclization step. And then fraudulently sends over the client's copy of the spectra of the correct isomer, claiming he made the right stuff all along. This guy got promoted for his hard work. The CRO was fired and paid next to nothing for their efforts.

HUGE waste of our time and setback to meeting our milestone.

12. Giagan on June 22, 2006 12:53 PM writes...

Another waste of time having to do with crystallization, this one the fault of a couple of chemists--

While in grad school, a student and post-doc in my lab were working on a natural product. Near the end of the synthesis they obtained some beautiful white crystals. Incidentally, this was immediately after having installed a para-methoxybenzyl (PMB) protecting group using PMB-trichloroacetimidate. They grew a nice crystal and submitted it for X-ray analysis. The crystallographer called up to the lab a little later saying that he wasn't seeing a complex structure--rather something simple and trigonal. The structure turned out to be trichloroacetamide, the byproduct of the benzylation reaction. The synthetic chemists hadn't been clued in to its presence since it really has no 1H NMR signature.

At least the structure was simple and the crystallographer didn't lose too much time. I think all parties involved agreed that there really was no need to tell the boss about the incident!

13. hch on June 22, 2006 2:33 PM writes...

qetzal, apparently this happens with surprising frequency (a lot of labs use pLysS strains with their expression cells). a grad student in the crystallography lab one floor above me managed even to get synchrotron beam time and actually obtained the structure before realizing.

i inadvertantly mass spec'd lysozyme once after trying to make RNase H. luckily, i was able to ID it was by its exact MW. goes to show the amazing power of MS for proteomics assays, i suppose...

14. august on June 22, 2006 8:15 PM writes...

Giagan's comment reminds me of another grad-school crystallization fiasco, whereby a former colleague of mine in a carbohydrate group was heard wandering up and down the halls, crowing that he had just managed to crystallize an amino-functionalized disaccharide. The sugar in question over the previous 2 years had refused to form anything other than an amorphous foam, and his boss was of course itching to get a structure since this could prove a hypothesis he had been toying with for ~10 years. The final step of its synthesis was a Staudinger reduction of an azide to an amine. Of course, the grad student was bouncing off the walls and immediately sent his crystals off for analysis, without doing any of that boring characterization stuff, such as checking the NMR. While waiting for the XRD structure, he proclaimed to everyone within earshot that we'd read about it in JACS or Angew. Chemie soon. Of course, the crystallographer returned the best-looking structure of triphenylphosphine oxide that any of us had ever seen.

15. Derek Lowe on June 22, 2006 8:51 PM writes...

August, as soon as you mentioned the Staudinger reduction, I began to shiver, like someone watching a horror movie, saying "No! Not the perfect looking crystals! Don't do it! Aaaagh!"

16. Mr. Vent on June 23, 2006 7:58 AM writes...

I remember a Hit Optimization project, we had great chemical feasibility, and a very good EC50 as starting point. We made hundreds of compounds but could not found a SAR or real improvement. After a while we could do DLS-studies and found out our compounds aggregated like hell. A great optimization of a false positive.

In another project tests were done on a mutant receptor. After a year synthesizing the normal human receptor became available and all the compounds proved inactive on this receptor. The project is now in the caves of our archive under a large pile of dust, and no-one dares mentioning the name of this GPCR again.