Because of these several impediments, as well as other technical difficulties, you will find typos and occasional replication of text. In our humble opinion, typos show that the logs are of *live* chats, not canned interviews, and minimal editing of these logs has taken place.

Ronald Taylor's Home Page

[Ron_Taylor] Hello, folks. Glado be here.
[Cybling] Ron, you're a computational biologist and a software writer.
[Kimmo] Hello!
[Cybling] What the heck is computational biology?
[Ron_Taylor] It is a broad field.
[Ron_Taylor] A couple subfields:
[Ron_Taylor] protein folding and design and
[Ron_Taylor] sequence analysis ( gene prediction,
[Ron_Taylor] for example)
[Cybling] Okay...I'm still not clear on this Ron.
[Cybling] You use computers to fold proteins?
[Ron_Taylor] Yes, there are scientists that use computational models
[Ron_Taylor] to fold proteins from first principles. Also,
[Ron_Taylor] there are others that look for similaries in sequence to
[Ron_Taylor] proteins with structures that are already known, for guidance.
[Cybling] So you write the software that genetics researchers use?
[Ron_Taylor] Yes, sometimes. My particular subfield is gene expression analysis,
[Ron_Taylor] although I also work in other subfields.
[Cybling] Okay...Ron....I know about as much about "gene expression" as I do about football here...
[Cybling] what does gene expression mean?
[Ron_Taylor] Some background: you have probably heard of the Human
[Ron_Taylor] Genome Project, which finds the sequences of human genes.
[Cybling] Yes indeed.
[Ron_Taylor] Once that is done, then the sequences have to be combined
[Ron_Taylor] to actually yeild the precise seq for each gene.
[Ron_Taylor] And, following that, we want to determine what genes are
[Ron_Taylor] active in a particular cell type at a particular time`
[Ron_Taylor] That is where gene expression analysis comes in.
[Cybling] Okay...so we're talking about a really, really, really massive database here....and something more as well?
[Ron_Taylor] Yes, what we try to do is measure the gene exp level of
[Ron_Taylor] thousands of genes at once, and then try to find patterns
[Ron_Taylor] within that set of data that, for example, might characterize
[Ron_Taylor] cancerous cells from normal cells.
[Cybling] And PERL just wouldn't do the trick here. This is not a flat text file.
[Cybling] What programming language do you use?
[Ron_Taylor] No, typically a relational database is used to store the data.
[Ron_Taylor] Perl is actually used as the glue for some of the tools. Jave
[Cybling] LOL...I suddenly fell much less inadequate here.
[Ron_Taylor] Whoops - I mean Java is a very popular language.
[Ron_Taylor] That is because a lot of our tools in the comp bio community
[Ron_Taylor] are delivered over the web.
[Cybling] Cool.
[Cybling] YOu're working in Colorado right now, right?
[Cybling] Are most of the folks working on this project here in the states then?
[Ron_Taylor] Yes, that's right. I moved to the U of Colorado in June from NIH.
[Ron_Taylor] A lot of comp bio is done in the States, but there are also
[Ron_Taylor] active communities in Europe and Japan.
[Cybling] We're talking about some truly serious number crunching here...what types of computers do you guys need to sort through all this data? (Imagining Collossuss)
[Ron_Taylor] Well, for example: several people have told me that Celera,
[Ron_Taylor] the private company that is sequencing the human genome on
[Ron_Taylor] its own, uses a computer farm that rivals the computer power
[Ron_Taylor] used by the National Security Agency.
[Horus] :)
[Ron_Taylor] However, the TYPICAL biology lab simply has a couple PCs
[Horus] hehe
[Ron_Taylor] or Macs, and analysis software has to designed to run on
[Ron_Taylor] what is available to such people, if we desire to help hte
[Ron_Taylor] average bench scientist.
[Cybling] Okay, Ron, can we follow the path of a gene from the Genome Project to you to the database? How does it typically run?
[Cybling] If you have questions for our guest, please just ask. But please, let the guest answer one before asking another. Thanks!
[Ron_Taylor] The first step: fitting the sequence fragments together into
[Ron_Taylor] what is known as a "contig". You have to do this because you
[Ron_Taylor] can only sequence 400-500 bases at a time, and a human
[Ron_Taylor] gene can span thousands of bases (with a lot of non-coding
[Ron_Taylor] "introns" in-between the coding parts, which adds even more
[Ron_Taylor] bases to what you have to hook together).
[Ron_Taylor] Second: you have to extract the coding subseqs from the intervening
[Ron_Taylor] introns very precisely and combined them to get the message
[Ron_Taylor] that actually produces the protein.
[Cybling] So...hmmm.....you're translating human programming, which is what this is sounding a lot like, into something a computer can read?
[Cybling] And sort through.
[Ron_Taylor] No, not exactly. We are trying to find the precise set of
[Ron_Taylor] bases that codes for a given peptide.
[Ron_Taylor] Even once that is done, there are complications.
[Ron_Taylor] One: there are what are called "post-translational" modifications
[Ron_Taylor] to the protein, quite often, None of those modifications
[Ron_Taylor] show up in the DNA or mRNA that codes for the peptide.
[Ron_Taylor] This definitely complicates the problem of determining
[Ron_Taylor] just what peptides are active in a given cell.
[Cybling] Ron...Are there more detailed informational pages linked from your website?
[Ron_Taylor] Well, you can start by going to http://www.ncbi.nlm.nih.gov
[Ron_Taylor] That is the home page of the fed gov agency that is most
[Ron_Taylor] involved in computational biology, and will have links to
[Ron_Taylor] a lot of other sites.
[Cybling] Ron...you're part of the science contingent here at the convention... have you guys been throwing any wild parties at all? This convention hasn't been all work for you has it?
[Ron_Taylor] Nope, sad to say, no wild parties that I know of.
[Ron_Taylor] But I have been having fun.
[Cybling] Oh, that's too bad. You are staying in town for the dead dog parties tonight though aren't you?
[Ron_Taylor] Yes, I'll be here till tomorrow morn.
[Cybling] Oh good...then perhaps you get to see at least some of the things a lot of folks come here for.
[Cybling] And have you been trapped in the dealer's room yet?
[Ron_Taylor] I wouldn't say trapped .. I have spent more $ than I budgetted for.
[Cybling] I think everyone has. One friend told me she spent what she budgetted...
[Cybling] then budgeted more...
[Cybling] and decided this is lot like Vegas, and she already has here tickett...
[Cybling] home, so what the hell.
[Ron_Taylor] Yes, I think I'll enjoy the books I bought. So what the heck.
[Cybling] Ron...I'm going to let you get out of here and go back to enjoying this convention, as a fan.
* Cybling undos duct tape
[Cybling] Okay, your free to go! LOL.
[Cybling] Thanks so much Ron.
[Ron_Taylor] Ok, goodbye folks.
[RedRaptor] by Ron, I found what I did see interesting