Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!cwjcc!gatech!mcnc!duke!dukempd!crown
From: crown@dukempd.UUCP (Rick Crownover)
Newsgroups: sci.bio
Subject: Re: Chromatin Packing
Summary: Chromatin Packing
Keywords: Chromatin
Message-ID: <709@dukempd.UUCP>
Date: 29 Aug 88 01:57:27 GMT
References: <707@dukempd.UUCP> <3006@boulder.Colorado.EDU>
Organization: Duke University Physics Dept.; Durham, N.C.
Lines: 28



                                                         The experiment I had in mind would have required synthesizing
a chain of known (and hopefully repetitive) sequence.  This chain
would then be placed in solution with the histones and allowed to
pack up (Albert's text suggested that this would occur spontaneously).
     After packing, a small molecule with ends capable of bonding to
specific AA's (which have been included in the original chain at
known intervals) would be introduced in a quantity large enough to
saturate the appropriate sites on the chain.  Picture this second
molecule as forming "bridges" between two points when possible, or
attaching like a branch on a tree when it can't reach a second
suitable site.
     Now denature the packed chromatin and cleave the chain with a
bond specific 'something-ase' which will break the chain -- again
in known positions along the chain.  At this point a determination
of molecular weights for the cleavage residues (terminology???)
would be performed.  By comparing the experimental weights with a
statistical model based expectations for the two possible structures,
it should be possible to select one or the other.
     If the chemistry is up to this,  the statistical model is easy.
X-ray crystallography is probably simpler though... and if its been
done...  Anyway, that was the idea.  Next.

-- 
	Rick Crownover				1-919-684-8279 
	Duke University Dept. of Physics	crown@dukempd.uucp
	Durham, N.C.      27706			mcnc!duke!dukempd!crown