20 January 1953
Dr. Jerry Donohue
The holidays and various activities have kept me from answering your letter of 16
December until now.
I asked Yakel to check over the calculations of the radial distribution functions
for the a helix, and he has done so. He has reported to me that the inter-atomic
distances that were used seem to agree with those that he has calculated, to within
about 0.1 A. The coordinates that were used in the calculations are, in fact, not
those given in our published table (which refers to a 3.6—residue helix), but those
obtained about a year earlier. They correspond to a 3.67—residue helix - the helix
with 11 residues in 3 turns.
I think that the smoothing function that we used was one that I devised on the basis
of information that you gave me. It involved spreading each point out over about
0.5 A to either side. I think that it is in between yours for B = 2 and B = 4.
I sent details about our smoothing function to Riley, and it is likely that it does
not differ much from the one that he used. It is not clear to me whether the difference
between our calculated curves and yours (and Riley's) is due to a difference in the
coordinates, or to something else. Perhaps we missed the double peak at 5 A for β
carbon 2 because of an error in one of our points.
Yakel, Schatz, and Rack have completed their work on the cylindrical distribution
function. They have evaluated the cylindrical distribution function for a keratin
and also for collagen from the experimental fiber diagrams, and they have evaluated
the cylindrical distribution function for the a helix, using the coordinates for the
3.67-residue helix as the starting point. There is only rough agreement between
the theoretical function and the experimental function for a keratin. There is of
course, disagreement for collagen.
I expect that you have seen our article in NATURE, on the a-keratin proteins. We
have made more detailed calculations, but have not yet written the material up for
publication in the PROCEEDINGS.
Did I tell you that Corey and I have sent a paper on the structure of nucleic acids
off to the PROCEEDINGS for publication? It will appear In the February issue.
As to Carlisle, I went over his material on ribonuclease with him and Bernal this
summer. I reached the conclusion that the Fourier diagrams that he obtains, and
adduces as evidence against a helixes and in favor of a ribbon structure, are pure
artifacts - that there is no justification for assuming the signs of Fourier terms
in the way that he does. There is, of course, the possibility that
I misunderstood the argument that he presented, but I do not think that I did.
I hope to see you in Cambridge about 6 April - I plan to stop off for the weekend
while on the way to Brussels.
With best regards, I am