MUD transcript: 10-12-99
Primary Structure of Proteins
Course room(s): PPS
22:59:20 LarryT connects.
22:59:20
LarryT PPS Telnet 0s 0s
22:59:24 DavidM connects.
22:59:24
LarryT PPS Telnet 4s 4s
DavidM PPS HTTP 0s 0s
23:00:22 ClareS connects.
23:00:22
LarryT PPS Telnet 1m 2s 1m 2s
DavidM PPS HTTP 58s 38s
ClareS PPS HTTP 0s 0s
23:00:30 ClareS says "Hello"
23:00:33 RamananS connects.
23:00:33
LarryT PPS Telnet 1m 13s 6s
DavidM PPS HTTP 1m 9s 49s
ClareS PPS HTTP 11s 3s
RamananS PPS HTTP 0s 0s
23:00:54 ClareS says "sorry if I'm a couple of minutes late - but that's not unusual"
23:01:17 ClareS says "I'll wait a few minutes before starting properly to give others a chance to arrive"
23:03:55 LesleyM connects.
23:03:55
LarryT PPS Telnet 4m 35s 10s
Wizard Void Telnet 3m 1s 3m 1s
DavidM PPS HTTP 4m 31s 25s
ClareS PPS HTTP 3m 33s 14s
RamananS PPS HTTP 3m 22s 18s
LesleyM PPS HTTP 0s 0s
23:05:37 ClareS says "Hi - welcome! I'd like to start the meeting now... by thanking you all for turning up"
23:06:18 LesleyM says "hi Clare, I think I' connected."
23:06:41 DavidM says "Hello, Clare"
23:07:26 ClareS says "Hi again, sorry I lost the connection for a minute then"
23:07:42 ClareS says "how are you getting on with the course so far?"
23:07:43 LesleyM says "Are we supposed to be discussing any thing in particular this session?"
23:08:16 ClareS says (to lesleym) "only the first two sections of the course material!"
23:08:50 ClareS says (to lesleym) "they're quite straightforward ones, so this is also another opportunity for you to learn about the MUD"
23:09:15 ClareS says "but if any of you have any particular questions you'd like to ask, go ahead..."
23:09:41 LesleyM says "well that's quite a bit. I have a question re Implications of Primary Structure Material. What is meant by "resonable allowance for insertions and deletions"?"
23:10:17 ClareS says (to lesleym) "could you post the URL, please? I'd like to look at the context"
23:11:41 LesleyM says (to say) "cryst.bbk.ac.uk/ppscore/section2/implications.shtml"
23:12:34 ClareS says (to lesleym) "thanks!"
23:14:21 ClareS says (to lesleym) "I've found that section now... it means that, if you have two proteins that are (very approx.) 15-25% identical..."
23:15:04 ClareS says "... so that you are not sure from identity alone whether they are homologous (evolutionarily related) or not..."
23:15:41 ClareS says "... they are more likely to be related if there are few insertions or deletions in one protein relative to the other"
23:15:52 RamananS disconnects.
23:15:52
LarryT PPS Telnet 16m 32s 5s
Wizard Void Telnet 14m 58s 1m 40s
DavidM PPS HTTP 16m 28s 9m 11s
ClareS PPS HTTP 15m 30s 11s
LesleyM PPS HTTP 11m 57s 4m 11s
23:15:59 ClareS says "you might have two proteins of very different lengths where one section is quite similar"
23:16:30 ClareS says "in that case you may well have a conserved domain but the proteins wouldn't (necessarily) have the same overall structure"
23:16:38 ClareS says "does that make sense?"
23:16:57 ClareS looks at david for confirmation
23:19:06 RamananS connects.
23:19:06
LarryT PPS Telnet 19m 46s 3m 19s
Wizard Void Telnet 18m 12s 13s
DavidM PPS HTTP 19m 42s 12m 25s
ClareS PPS HTTP 18m 44s 1m 55s
RamananS PPS HTTP 0s 0s
LesleyM PPS HTTP 15m 11s 7m 25s
23:19:13 LesleyM says "I think it does. Let me see - two proteins with longish identical sections would be thought of as homologous even if there were then some significant insertions/deletions upto 20%.?"
23:19:55 ClareS says "not quite... I don't think I've explained it very clearly"
23:20:11 ClareS says "you have 2 proteins about 20% identical..."
23:20:48 ClareS says "if the similarity is spread over the whole length of the protein, with few long gaps, they are (most) likely to be (distant) homologs"
23:21:23 ClareS says "if one or more sections are well conserved but there are long gaps with few identities..."
23:21:51 ClareS says "... I'd say it was less likely that the proteins were directly related tho' they will certainly share one or more structural or functional domains"
23:22:17 ClareS says "(e.g. an immunoglobulin domain, or a tyrosine kinase domain"
23:22:28 ClareS says "is that clearer?"
23:22:38 ClareS must remember to close brackets
23:23:32 LesleyM says "I think I get the idea but may be fixing too much on the 20% value - where does that come from? Is it empirical, statistical or what?"
23:23:51 DavidM says "If the 15-20% identity is uniformly spread then there is less likely to be an evolutionary relationship than if there are insertions of lower homology and longer regions of higher homology"
23:24:27 DavidM says "That is what Clare is saying, I think!"
23:24:45 ClareS says "the 20% is empirical -- it just turns out to be a useful cut-off. It won't apply in all situations"
23:25:17 ClareS says (to davidm) "you've explained it clearer than I did"
23:25:39 LesleyM says "OK - I can see that. When might it not apply?"
23:25:52 ClareS says "tho' proteins which are 20% identical with uniform identity are quite likely to be distantly related"
23:26:37 LesleyM says "sorry to hog all your time. I am clearer now."
23:26:43 DavidM says "Ther is a paper by Mike Sternberg in J Mol Biol, 1999 which looks at homologous sequences of low identity. I haven't got the reference here. Clare knows the paper."
23:26:48 ClareS says "you take a family like the globins -- some globins are only about 15% identical but this is a very clear evolutionarily related family"
23:26:48 RamananS says "thanx wizard pages;Learning a lot already watching you guys talk"
23:27:12 ClareS says "some proteins with that degree of identity are not related at all"
23:27:42 ClareS says "particularly proteins that have slightly odd compositions, e.g. coiled coils contain a high proportion of certain amino acids"
23:27:57 ClareS says "you will then get quite high identities "by chance""
23:28:08 LesleyM says "Is the J Mo Biol paper useful in illustrating these differences?"
23:28:31 ClareS is looking up the full reference in medline
23:29:10 ClareS says "you can probably trust mike sternberg more than either david or me on this, he's studied this problem for many years"
23:29:29 DavidM says "I will look at the paper again and if it is relevant we can email the reference"
23:29:36 LesleyM says "will look it up"
23:30:46 LesleyM says "another question - how much should we know/be able to relate re chemical techniques for sequencing?"
23:32:17 ClareS says "The paper David was referring to must be J Mol Biol. 1999 Nov 12;293(5):1257-1271"
23:32:43 ClareS says "the title is "Benchmarking PSI-BLAST in Genome Annotation""
23:33:07 LesleyM says "thank you"
23:33:11 ClareS says "http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=10547299&form=6&db=m&Dopt=b"
23:33:34 ClareS says "sorry, I don't think that that link is fully clickable"
23:34:02 DavidM says "Mike looks at the ability of PSI-BLAST to pick up homologies at low sequence identity"
23:34:16 ClareS says "there are statistical tests for working out whether two proteins are similar enough to be definitely related, that take sequence composition into account"
23:34:57 ClareS says "is that clear (more or less)?"
23:35:17 LesleyM says "is this PSI_BLAST statistical?"
23:36:03 ClareS says (to lesleym) "it uses statistical techniques to list sequences in order of probability of homology (if that makes sense?)"
23:36:34 LesleyM says "yes, I think it's clear. Many thanks."
23:36:37 ClareS says "all programs for searching databases with sequences do that"
23:37:21 ClareS says "the difference with PSI-blast is that it is iterative - it creates an alignment of related sequences from the first search"
23:37:34 ClareS says "and goes on to re-search the database with that alignment"
23:38:45 DavidM says "We will come back to sequence alignment in the bioinformatics part of the course"
23:39:23 ClareS says (to davidm) "I don't think Terri's exercise includes PSI-blast: can you remember whether it does?"
23:39:34 LesleyM says "I leave it for now - need to look at that paper"
23:40:07 ClareS says "bioinformatics (mostly sequence alignment) is covered quite extensively in section 5 of the course"
23:40:25 ClareS says "you'll come on to that in late January - the second section of the new year"
23:40:26 RamananS says "would the course cover how these databases work in more detail at some stage in 99/00"
23:40:39 ClareS got there before Ramanan ;)
23:42:35 ClareS just lost the connection for a couple of seconds again
23:43:01 RamananS says "this is great ;can I diturb and ask questions on problems telnetting"
23:44:22 ClareS says (to ramanans) "of course - go ahead"
23:44:29 RamananS says "I figured it out ;the port no. was wrong sorry!"
23:44:54 ClareS says "it is *very* important to get the port no. right ;)"
23:45:21 ClareS says ".. and to remember that the port no. for telnet to the MUD is not the same as for the HTTP interface"
23:45:53 ClareS says "bioinformatics is probably the most popular part of the course, at least at the moment"
23:46:08 LesleyM says "please jump in when you want Ramanan"
23:46:52 RamananS says "Sorry I've been not upto it totally.But will catch up by the next MUD session."
23:46:58 DavidM says "The port numbers over 1024 are used for special purposes such as MUDs"
23:46:58 LesleyM says "about the chemical method of sequencing - how much is it advisable to know?"
23:47:43 DavidM says "The low numbers are all standardized. So 80 is for HTTP (eg browsers) and 23 is for telnet"
23:47:56 ClareS says (to lesleym) "sorry, I didn't answer that question earlier - it arrived in the middle of our long discussion of distant homology"
23:48:21 RamananS says "i haven't looked at the released material in detail yet;can you recommend a good e-journal searching program"
23:48:37 ClareS says "it is not necessary for you to know a great deal about chemical sequencing - in fact I can't remember a single mention of it in the course"
23:48:45 ClareS says (to davidm) "can you?"
23:49:18 DavidM says "We are developing another Adv Cert with that sort of thing in it"
23:50:38 LesleyM says "there is brief mention in 3.5 sequencing (talk of 6N HCl...) for cleavages.."
23:50:41 ClareS says "if you complete PPS and then that new course (probably not available until some time in 2001) we hope that you may be eligible for an M.Sc."
23:50:52 ClareS says "we are currently working out most of the details"
23:51:27 ClareS says (to lesleym) "you don't need to know about how the sequencing is performed, tho'"
23:52:20 LesleyM says "the new cert also sounds quite appealing (slushy stuff for a change)"
23:52:21 RamananS says "Isn't general sequencing totally automated yet?"
23:52:21 ClareS says (to ramanans) "do you know BioMedNet?"
23:52:33 ClareS says "http://www.biomednet.com"
23:52:56 ClareS says "if any of you are not yet members of biomednet it's well worth joining - and it's free"
23:53:34 ClareS says "you get access to full evaluated Medline, and if *your institution* has bought the paper version of a journal you may be able to see the full text online for free"
23:53:57 ClareS says "even if you don't have access you can buy an individual article for a few dollars with a credit card"
23:54:05 RamananS says "are their any other servers/resources that we have access to with the bbk passwords"
23:54:28 ClareS says "not with the PPS passwords"
23:55:33 ClareS says (to lesleym) "it's all the theory behind the techniques, we can't teach practical mol. biol. online (yet ;)"
23:55:50 DavidM says "I was thinking about that. We will have to check this with our librarian"
23:56:10 RamananS says "I am sorry what I meant was ,do we have free access too any other resources that one would be expected to pay otherwise"
23:56:27 ClareS says (to ramanans) "most protein sequences these days are obtained by translating gene sequences - obviously gene sequencing is now automatic and very fast"
23:57:07 ClareS says "the only difficulty is that the programs available for predicting gene locations are still quite crude"
23:57:23 LesleyM says (to ClareS) "yes that would be a problem!"
23:57:47 ClareS says "david was saying that we will check with Birkbeck's library staff whether you can have access to the online services provided via Birkbeck's library"
00:01:00 LesleyM says (to ClareS) "I have purchased Voek & Voek, plus Branden a& Tooze. Should this suffice for basic texts?"
00:01:27 ClareS says (to lesleym) "Yes, I should think so!"
00:01:53 ClareS says (to lesleym) "(assuming that you mean Voet & Voet - I've never heard of a textbook author called Voek ;)"
00:02:09 DavidM says "There is Voet, Voet and Pratt which has a CD-ROM with it and is very good"
00:03:08 LesleyM says (to ClareS) "yes that's the one although sans CD (I'over CDed anyway for now)"
00:03:29 ClareS says "that is the newer edition of Voet & Voet - they acquired another co-author when they brought out the CD"
00:04:11 ClareS says "I think I must go quite soon: I am beginning to fall asleep at the keyboard. Are there any final questions?"
00:04:59 DavidM says "Does everybody know that the Birkbeck network is going to be down this weekend"
00:05:01 RamananS says "what's Prat all about?Strucural biology....?"
00:05:03 ClareS looks round expectantly
00:05:17 LesleyM says "I will leave you all to your slumbers now - sleep well and many thanks!"
00:05:35 ClareS says (to ramanans) "I'm not quite sure what you mean?"
00:06:03 ClareS says "yes, David has made a good point: please remember to log off when you leave the MUD and don't try too hard to look at the course over the weekend"
00:06:12 ClareS says "our server is being upgraded"
00:06:14 DavidM says "Quite alot of V,V and P is about structural biology in the biochemical context"
00:07:06 RamananS says "I get it good old voet ,voet and Pratt.I have nop more stupid Qs;Good night everyone"
00:07:29 LesleyM waves goodbye
00:08:03 ClareS yawns
00:08:05 DavidM waves
00:08:07 LesleyM disconnects.
00:08:07
LarryT PPS Telnet 1h 8m 23s
Wizard Void Telnet 1h 7m 14m 44s
DavidM PPS HTTP 1h 8m 2s
ClareS PPS HTTP 1h 7m 4s
RamananS PPS HTTP 49m 1s 1m 1s
00:08:07 LesleyM connects.
00:08:07
LarryT PPS Telnet 1h 8m 23s
Wizard Void Telnet 1h 7m 14m 44s
DavidM PPS HTTP 1h 8m 2s
ClareS PPS HTTP 1h 7m 4s
RamananS PPS HTTP 49m 1s 1m 1s
LesleyM PPS HTTP 0s 0s
00:08:12 LarryT disconnects.
00:08:12
Wizard Void Telnet 1h 7m 14m 49s
DavidM PPS HTTP 1h 8m 7s
ClareS PPS HTTP 1h 7m 9s
RamananS PPS HTTP 49m 6s 1m 6s
LesleyM PPS HTTP 5s 5s
00:08:23 ClareS says (to davidm) "thank you for helping me out with some of the questions ;)"
00:08:47 DavidM says "Thank you Clare"
00:09:02 DavidM says "Bye"
00:09:15 ClareS waves goodnight
00:09:28 ClareS disconnects.
