home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.space
- Path: sparky!uunet!sun-barr!ames!nsisrv!stars.gsfc.nasa.gov!bhill
- From: bhill@stars.gsfc.nasa.gov (Robert S. Hill)
- Subject: Re: ET's, life in space
- Message-ID: <30JUL199214461107@stars.gsfc.nasa.gov>
- News-Software: VAX/VMS VNEWS 1.4-b1
- Sender: usenet@nsisrv.gsfc.nasa.gov (Usenet)
- Nntp-Posting-Host: stars.gsfc.nasa.gov
- Organization: Hughes STX Corp./NASA Goddard Space Flight Center
- References: <1992Jul29.225958.20821@cs.cmu.edu> <rwallace.712484402@unix1.tcd.ie>
- Distribution: sci
- Date: Thu, 30 Jul 1992 19:46:00 GMT
- Lines: 49
-
- >In <1992Jul29.225958.20821@cs.cmu.edu> 18084TM@msu.edu (Tom) writes:
- >>>>Given that for evolution of life to start, a simple living organism must
- >>>>come together from amino acids etc. by accident; and that for any
- >>Not accident. Think in statistics. Given umpteen billion gallons
- >>of warm, super gunky sea, what are the possibilities that any one
- >>group of amino acids could form self-replicating protiens...
-
- In article <rwallace.712484402@unix1.tcd.ie>,
- rwallace@unix1.tcd.ie (russell wallace) writes...
- >I *am* thinking in statistics. If the smallest possible self-replicator
- >is 1000 amino acids in size, and there is only one self-replicator of
- >that size, and supposing there are 20 possible amino acids, then the
- >odds against any 1000-amino acid chain being a self-replicator are
- >20^1000 to 1. Against this statistic, the total number of molecules that
- >can have been formed in the visible universe since the beginning of
- >time is insignificant.
-
- Any time I see the phrase `the odds against X,' a little alarm bell
- goes off in my head. Let's postulate a situation and do the combinatorics.
-
- Suppose the `reaction vessel' contains 1E6 molecules of each of 20
- amino acids. Suppose that our magic polypeptide contains 1000 amino
- acids, 50 of each kind for simplicity. Let C(N,n) mean the combinations
- in N objects taken n at a time (since it's a little hard to make double-
- height parentheses on my terminal). ^ is exponentiation operator.
-
- Prob(magic molecule) = C(1E6,50)^20 / C(2E7,1000)
-
- I think I have this formula right. Probability mavens, please correct
- me. The denominator is the number of 1000-link polypeptides that can be
- selected from our reaction vessel. The numerator is the number of them
- that are this one kind of active molecule. I evaluate this probability
- to be 1E-23 _not_ 20^(-1000). (First time I've used logs to do a
- computation since high school.)
-
- There are about 1E17 seconds in 2 billion years (oops, I mean 2E9 years
- for non-USAns). So if you have one chain spontaneously form and fall
- apart per second (arbitrary assumption), that gives you 1E17
- `statistical trials' in two billion years. If you have 1E6 of the
- `reaction vessels' such as I've described, you have a good shot at
- forming your molecule.
-
- Of course, I haven't _proved_ a darned thing, but I think that the
- `horrendous odds against' argument is pretty shaky.
-
- - Bob Hill
- ---------- ---------- ---------- ---------- ---------- ---------- ----------
- bhill@stars.gsfc.nasa.gov ***MY individual opinions ONLY***
- ---------- ---------- ---------- ---------- ---------- ---------- ----------
-