From: Derek Gatherer (email@example.com)
Date: Tue 06 Dec 2005 - 09:02:11 GMT
At 20:40 05/12/2005, Dace wrote:
>Here's the relevant passage, from the bottom of page 213:
>"The serial assay was redesigned as follows. A cell line was grown from
>small inocula in a culture medium containing no selection agent. At each
>passage, one or more cell samples were withdrawn and seeded into separate
>flasks (instead of a depleted monolayer) containing the selection medium.
>In these flasks, because of selection pressure, the wt cells died while
>drug-resistant cells gave rise to visible colonies. Mutation frequency at
>each passage level of the cell line was then calculated from the total
>number of colonies divided by the number of cells assayed for resistance
>and corrected for cloning efficiency."
>What Hill has demonstrated is that when a colony of cells is subjected to a
>toxin and develops resistance to this toxin (over the course of
>generations), the same mutation enabling this resistance will also begin
>appearing in a closely related but physically separated culture. This
>indicates a nonrandom pattern of mutations in direct contradiction to
No, what it demonstrates is that putting fresh cells into a flask
which has previously been used, does odd things to the cells, in this
case increasing the mutation rate. What Hill compares is:
1) Cells from the parent stock, seeded into clean flasks.
2) Cells from the parent stock, seeded into serial culture (the same
flask into which the previous seeding went)
Let me try a metaphor on you. Imagine I was trying to investigate
the effect of chocolate on prisoners. One set of prisoners are
rehoused every night in clean cells (no pun intended) and not given
any chocolate. The other group are serially replaced into the same
cell containing the dead bodies of the former inmates, and fed
chocolate. Surprisingly they suffer from psychiatric symptoms more
than the other group of prisoners. Would you therefore conclude that
chocolate has an adverse effect on mental health?
>Incidentally, Hill has responded to your initial complaint, namely that if
>his results were true, researchers would already have spotted this
>Dear Ted Dace,
>Researchers don't see nonrandom mutations in unexposed cultures because
>they don't look for them in a good way. To my best knowledge, no one ever
>tried a serial assay though such assays are known from 1986 (see below).
Yes, precisely for the reason (see the prisoners above) that serial
assays are uncontrollable (well, there is one possible control - not
a particularly good one - but I'll leave you to try to think of what it is).
>Unlike a standard procedure the serial assay "extends across several
>passages". This may last for several months, showing that during this time
>period the cell population contains no resistant mutants.Then at further
>passages these mutants are found at small but increasing numbers. The
>absence of such mutants at low passages and their occurrence at high
>passage levels, taken together, provide evidence of nonrandomness.
Even worse. If you've really got to wait "several months" to see a
marginal effect, then it is really difficult to control for all the
other sources of variation that might creep in during that time.
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
This archive was generated by hypermail 2.1.5 : Tue 06 Dec 2005 - 09:39:45 GMT