Among the odder things to do on Maundy Thursday

is to write a Haskell program to enumerate polyominos. There are 369 such consisting of eight tiles.

To my slight disappointment, even when compiled with ghc, it is about 12% slower than the Python equivalent. And the Python draws the tiles out to a .PNG file, whilst I've no idea how to get the Haskell to do that.

This is the software equivalent of a CRIT request: could I ask Haskellian or Pythonesque readers to have a look at the code and point out where I'm committing staggering solecisms of style and speed?

OK, probably many of them will be at Eastercon as they read this, as will I be tomorrow.

Reinforced magic, upon stilts

If I hear at Eastercon anything as purely sense-of-wonder-inducing as the collection of articles below, I'll be amazed.

There's a research group - relevant names are Floyd Romesberg and Peter Schultz - at the Scripps Research Institute who are modifying the absolutely fundamental genetic cycle. They've managed, in separate experiments, to introduce an extra mostly-correctly-replicated nucleoside into the DNA language ([1-deoxyribose-3-fluoro]benzene, bracketting incorrectly in the hope of explaining where the substitutents are; it's not completely unlike deoxycytosine, and here is a nice reference for how the bases are stuck to the sugars on DNA), and to introduce an extra amino acid, with a *four*-letter codon to code for it, into the protein synthesis process.

This is more fundamental than inserting an extra letter into the English alphabet, because languages aren't active in the way that DNA is: the analogy would have to be Hermione discovering the new possibilities of magic words containing sounds that can't be represented in the international phonetic alphabet; it's rather more like making some bytes inside a working computer nine bits long.

At the moment, the extra amino-acid works in vivo, and the extra nucleotide in vitro. I suspect they'll be made to work together in some sufficiently exotic bacteria within a decade, shortly after which three people from the group will get a free trip to Stockholm.

The techniques used are as exotic as the goal; at current state of the chemical art, you can't design synthetase enzymes, polymerases or transfer RNAs, so you have to set up a bacterium and an environment such that there's a strong selection pressure for things with the right property to evolve.

http://www.scripps.edu/newsandviews/e_20020909/print-romesberg.html describes Romesberg's work with a section about the extra nucleotide; the interestingly-titled Efforts Toward the Expansion of the Genetic Alphabet has an indication of what they're doing.

http://www.scripps.edu/newsandviews/e_20040517/onpress.html is the extra-codon press release

http://www.pnas.org/cgi/reprint/0401517101v1 is the extra-codon paper

Many thanks to Derek Lowe for pointing me at the extended-codons work, and explaining patiently what 3FB means. I'm not quite sure why 3FB hit the scientific press now, the extra-nucleotide work has been going on for at least five years now.

Things fit together

I hadn't realised that the A in ATP, adenosine triphosphate, the universal fuel of cellular processes, was the same chemical as the A in the ACGT genetic alphabet. Had you?