ClareS says "testing recorder and tape"
ClareS turns the ClareS_recorder off.
ClareS turns the ClareS_recorder on.
ClareS . o O ( connection problems? )
ClareS says "there seem to be a few problems with the line tonight"
Henryb says "Either that or he didn't like the decor in the room!"
ClareS says "but we can press on anyway - I hope ;)"
(Gmocz has connected.)
David.cavanaugh finds his way in.
ClareS says "Hi, welcome - we've only just started (connection problems)"
ClareS says "may I start by introducing Henry Brzeski (Henryb) who wrote the Elements of Transcription material we use in this chapter"
Gmocz says " It looks like I got disconnected, but I am luckily back"
David.cavanaugh says "Hi all"
Henryb takes a bow
Cantellow [guest] says "hello"
ClareS says "anyone like to start with a question?"
Gmocz says " He" Hello henry"
Gmocz says " In general, how do the various DNA binding proteins recognize
their SPECIFIC binding sites? By a particular H-bonding pattern in the DNA,
or by some irregularities in the double helix, or by a special shape,
geometry or 3D structure of DNA a
t the binding site, or just by plain spatial complimentarity? What
parameters influence the specificity of DNA-protein interactions? How can a
common HTH, Zn-finger or bZip motif be specific for many different sites? Is
there a universal recognition code
or is there a AA side chain:DNA base-pair recognition code?"
Henryb says "Hi everyone"
ClareS [to gmocz] good question - very complex! It could last all meeting
if we want it to
Henryb [to gmocz] Can't we start on the simple stuff first!!
ClareS says "I should point out that there is a whole section on the *structures* of DNA / protein recognition sites later in the course"
ClareS says "so we will be going into this topic in a lot more depth then"
Henryb says "Unfortunately, I'm at home and dialling in so I do not have
access to my reprints but in essence interactions take place fairly
specifically and involve defined amino acid base interactions"
Gmocz [to HenryB"]: What statistical measures can be used to find ORFs?
Also, if all 6 frames can be translated, is the longest uninterrupted
stretch the most likely candidate for an ORF?
ClareS . o O ( I wonder if the arachnid is one of us )
ClareS [to gmocz] in simple terms the longest uninterrupted stretch *is*
the most likely ORF candidate...
Cantellow [guest] . o O (I'm a bit out of my depth here...)
ClareS [to gmocz] there are a number of programs for ORF recognition which
also use statistical measures
ClareS [to gmocz] e.g. base composition
Henryb says "The recognition sequence in the protein is different for each
abse sequence recognised and may involve hydrophobic interactions eg methyl
group of thymidine interacting with the aromatic ring of tyrosine"
ClareS [to Cantellow] don't worry about it - this is quite complex stuff
Cantellow [guest] [to ClareS] phew...
ClareS [to Cantellow] is there anything in particular that is unfamiliar to
you?
Gmocz says " : " Because of the redundant nature of the genetic code, a
protein sequence cannot always be translated back unambigously to a DNA
sequence. When such back translation is done, is a default standard codon
used for an AA, or the most frequent
ly occuring codon? Or, ideally, a codon usage frequency table is or should
be employed?"
ClareS says "back translation is by definition unreliable - it will only produce "a" possible DNA sequence not "the" definitive one"
Cantellow [guest] [to ClareS] My background in Biology is from studying 1st
year Medicine 5 years ago - so I covered a great deal of this in the
Biochemistry course... however, it's been a long time....
David.cavanaugh says "I have a question regarding Henry's statement
concerning a methyl group interacting with an aromatic ring. I would have
thought an alkyl to aromatic ring interaction to be quite weak."
ClareS says "again, this is species dependent - some species use some codons more frequently, depending on AT/GC ratio which varies greatly"
ClareS [to Cantellow] did you find the Cell Biology textbook and other
material we linked to helpful?
Henryb [to david.cavanaugh] weak it may be (I'm not going to make any
comments about that) but DNA-protein interactions are predominantly about
specificity
ClareS [to david.cavanaugh] any of these interactions alone is quite weak -
and a single hydrophobic interaction weaker than a h-bond. It's the sum of
all interactions that makes the specificity
Cantellow [guest] [to ClareS] yes - a good memory jogger.
ClareS is glad it was helpful ;)
Henryb [to david.cavanaugh] Don't forget that only about 5 bases are
onvolved in the specificity of binding so the strength of each specific
interaction will not be high
Gmocz while I was typing some answer came to my question, so I n Is a
hydrophobic interaction weaker alway than a H bond?
ClareS is trying to remember the figures...
Gmocz [to ClareS] This section was more difficult and complex than any of
the previous ones
David.cavanaugh says "A hydrophobic bond is much weaker than a hydrogen
bond, especially if there is a difference in the structure, say a simple
linear hydrocarbon to an aromatic compound"
ClareS says "it depends on the geometry of the H bond, some are very weak if they are long or the hydrogen angle is very far from 180deg"
ClareS [to gmocz] the first three sections are really introductory ones on
purpose
David.cavanaugh says "When proteins recognize DNA base sequences, is 5-7 bases sufficient to assure uniqueness of the sequence recognition ?"
ClareS [to gmocz] partly as everyone has to get to grips with the
technology at the start of the course and that takes time
Gmocz [to david.cavanaugh] This clarifies my understanding of hydrophobic
inyrtactions somewhat
Gmocz Sorry for the many typos. I am a bad typer.
ClareS [to gmocz] everyone makes mistakes, especially those of us on this
side of the planet at this time of night
David.cavanaugh says "Still early in the middle of the US"
Henryb [to david.cavanaugh] No but proteins don't rely on a single
recognition event. Sometimes binding sequences are palindromic and binding
proteins are dimeric
Gmocz Can I have another question? " RNA polymerase has two binding sites
for NTPs, the initiation site and the elongation site. The initiation site
has higher affinity for NTPs than the elongation site. Does this have a role
for the control of transcrip
tion in cells? Perhaps mRNA synthesis will not begin unless NTP
concentration is high enough for complete synthesis?
Henryb [to david.cavanaugh] Sometimes DNA binding proteins contain more
than one DNA binding sequence
ClareS [to henryb] the number of cases where a protein dimer recognises an
(approximately) palindromic base sequence is actually quite high
Henryb [to gmocz] I'm sorry but I'm not sure I understand your question
ClareS [to gmocz] I'm not sure that I do either - sorry
ClareS [to gmocz] could you expand on your last question, please?
Gmocz In other words because of the lower affinity, I guess, the synthesis
will not start until a treshold concentration of NTPs are present. This will
ensure that the synthersys will be complete and there will be no half
synthesized products. Is this co
rrect?
ClareS [to gmocz] it's a good question
answer without looking it up!
ClareS [to gmocz] sorry...
David.cavanaugh says "It used to be that folks spoke of Junk DNA. I get the
impression from the section 4 material, that this may not be the case. In
other words, there are the recognition sites and other structural needs
served by some sequences. Would
it be fair to assume then that we'll find there's very little junk after all
?"
Cantellow [guest] [to ClareS] I'm about 30 seconds behind everyone else...
ClareS [to david.cavanaugh] I am sure that a lot of what we now think of as
junk is not... introns, 3' and 5' untranslated regions will affect
transcription in different ways (some of which I guess we don't know yet)
Gmocz [to david.cavanaugh] I have not heard about junk DNA but if it does
not exist in sufficienly high concentration, that would explain my question
regarding the lower affinity at thne elongation site.
ClareS says "but I am also sure that we will never be able to figure out what (if anything) *all* of the so-called "junk" is for"
Cantellow [guest] says "Does DNA replication involve a lot of enery input?
How much would a eukaryote with little junk DNA be advantaged over one with
a lot of junk (from an evolutionary perspective)"
ClareS [to gmocz] it used to be thought that almost everything that wasn't
a gene was junk
Gmocz [to ClareS] Oh, that is something different then
ClareS [to Cantellow] I'm not sure
have *very* different genome sizes and still approximately the same number
of genes
ClareS says "there was an interesting discussion about mutations in
untranslated regions and genetic disorders in yesterday's meeting - it's on
the transcript"
Cantellow [guest] [to ClareS] does this apply to both pro- & eukaryotes?
ClareS [to Cantellow] eukaryotes - plants and fish in particular
Gmocz Is a mutation in a coding region always more "dangerous" than in a
non-coding region?
Henryb says "The tape also contains some comments about other functions for
introns in recombination and evolution of novel proteins by mixing and
matching protein domains"
ClareS says "there is a list of genome sizes for a large number of common genomes *somewhere" on the web"
ClareS says "the resource is known as DoGS - Database of Genome Sizes"
ClareS says "I can't remember the URL tho'"
Cantellow [guest] . o O (I'll go hunting for it sometime)
ClareS says "the puffer fish is an example of a eukaryote with a *very*
small genome - which is why it's used as a model organism. Many other types
of fish, some closely related, have genomes of similar sizes to other
vertebrates"
Cantellow [guest] gosh
Henryb [to gmocz] Mutations in coding regions, by definition, are more
'dangerous' than those in non-coding regions as many base changes in
non-coding regions will not have any effect on protein structure/function
ClareS says "they have similar gene numbers - the difference is all in what used to be called "junk DNA""
ClareS [to gmocz] mutations in non coding regions will not affect protein
structure / function but may well affect regulation
ClareS can't remember any numbers for genome sizes off hand
Gmocz [to HeneyB]: " But a mutation in a noncoding region might change the
organization of a gene which may have conseqvences. I guess this is much
more unlikely?
Henryb [to gmocz] Unless the mutation destroys a splice site leading to use
of cryptic splice sites and hence altering the mRNA sequence
ClareS says "so splice site mutations will sometimes prevent the translation of particular exons"
Henryb [to gmocz] Yes, alterations in non-coding regions will usually alter
protein levels by altering promoter strength or mRNA level
Gmocz [to henryb] Thank you. I just learned something.
ClareS is glad of that ;0
ClareS hopes that everyone learns plenty in BioMOOs ;)
Henryb [to ClareS] Don't forget the shift key ;0
Gmocz says " Why is DNA looping in transcriptional regulation common to
both prokariotes and eukariotes? Why is co-ordinated transcription
regulation unique to prokariotes? At what point of evolution was it given up
and why?"
ClareS says "I'd like to close the meeting in about 5 minutes - I'm getting tired and my typing's going (see above)"
David.cavanaugh says "It seems to me that in eukaryotic cells, that given
the number of introns which may occur, there possibly may be an information
protection mechanism associated with introns. Perhaps spreading the
information out over a longer distan
ce may serve as a protection against certain kinds of damage or information
loss."
ClareS [to david.cavanaugh] that's very likely true
ClareS [to gmocz] another question I'll have to look up...
Gmocz You mean, by looping spatially separated information can get close to
each other which might provide a certain way of regulation?
Cantellow [guest] [to david.cavanaugh] i.e. it reduces the odds of a
mutation occuring in a vital coding area?
ClareS [to gmocz] it's certainly true that in eukaryotes, some regulation
sites are a way away from transcription initiation sites. Is this what you
mean?
David.cavanaugh says "Yes, but I would assume there are other risks besides just point mutations."
Henryb [to gmocz] How can I answer all these? But let's try. Looping is
retained because it gives rise to greater specificity (our comments earlier
about length of recognition sequence), co-ordinated
transcription-translation regulation in prokaryotes on
ly because they don't have a nucleus. I wasn't around when evolution gave
these up so I can't answer your third question
Gmocz [to ClareS] more or less, yes
ClareS [to henryb] I guess it relates to the timescale of the evolution of
the nucleus...
ClareS [to david.cavanaugh]: undoubtedly (to "other risks besides point
mutations")
Henryb [to david.cavanaugh] but the larger a gene is the more chance of
deleting bits eg many mutations in the gene responsible for muscular
dystrophy are de novo deletions ie not genetically inherited
ClareS yawns
David.cavanaugh says "Much to think about here. Maybe we can get some good list discussion going."
ClareS [to david.cavanaugh] that's a good point on which to close the
formal discussion
ClareS [to david.cavanaugh] by all means mail the list with your
comments/questions from the meeting (& that goes for the others too, of
course)
ClareS says "thanks for your lively discussion, and thanks to Henry for his invaluable contribution"
David.cavanaugh says "Bye y'all"
Henryb says "I think the evolution aspect is very interesting and is now
becoming more understandable now that we have sequenced many organisms and
Comparative Genomics has arrived"
ClareS says "and goodnight!"
ClareS turns the ClareS_recorder off.