Wolfram
Hazard to Others
Posts: 133
Registered: 13-10-2003
Member Is Offline
Mood: No Mood
|
|
Bioinformatic tools for madscientists
Now Im planning to do a homepage containg a collection of bioinformatic tools for madscientists, with description of the tools for noobs. Do you know
of any good free webhotels without banners or maybee I could upload the page on sciencemadness server.. (if there is one)?
|
|
chemoleo
Biochemicus Energeticus
Posts: 3005
Registered: 23-7-2003
Location: England Germany
Member Is Offline
Mood: crystalline
|
|
As far as I know, free webhosts have banners/popups virtually by definition. There are some that don't have any, but these tend to be very slow.
I think you may just want to accept this.
One reasonably good one is free prohosting (check google for site address)
Nonetheless, could you please elaborate as to how to design that page? I fear it may be hard for any non-biologist/biochemist etc to really understand
what you get by aligning proteins/DNA, or by predicting folds and such. It can be explained of course.
But tell me, what is the layman supposed to obtain from this? If he doesn't really understand what he's just obtained?
I dont want to wreck your enthusiasm, but check www.expasy.ch
This is a vast collection of bioinformatic tools, for the professional scientist of course.
Make an amateur version out of this, and you will be a very popular man
Never Stop to Begin, and Never Begin to Stop...
Tolerance is good. But not with the intolerant! (Wilhelm Busch)
|
|
Polverone
Now celebrating 21 years of madness
Posts: 3186
Registered: 19-5-2002
Location: The Sunny Pacific Northwest
Member Is Offline
Mood: Waiting for spring
|
|
bioinformatics is "easy"
When I think of mad science, I always picture people exploring in a hands-on fashion. So designing a monster (or aligning some protein sequences) on
paper/silicon isn't enough. It's only once you've made a connection with the lab that it really becomes Mad Science. And though prices
are dropping, equipment and reagents in this field still look expensive.
It's hard for me to say if/how an amateur could use bioinformatics tools to make a lab connection. I'm taking a bioinformatics course right
now, and all we ever seem to discuss is biology (completely detached from experiments) or algorithms/software. So for example we've discussed why
structure tends to be conserved in proteins, and the computational complexity that makes heuristic methods necessary for alignment. But I don't
think we've ever touched on <i>why</i> protein sequnce alignment is important to people. It's as if biology never involves a
lab, or specific goals, from our discussions.
Astronomers can gaze at the heavens but they can't poke or prod them. Biologists can gaze <i>and</i> poke. I'd like to know if
bioinformatics ever enables "poking" in a way that Mad Scientists can appreciate.
I originally had a long ramble here about the role of software in science in general, and the quirks specific to scientific problems, but it
wasn't really on-topic.
PGP Key and corresponding e-mail address
|
|
madscientist
National Hazard
Posts: 962
Registered: 19-5-2002
Location: American Midwest
Member Is Offline
Mood: pyrophoric
|
|
I recommend using netfirms.com as a host. free.prohosting.com seems nice - until you start noticing how the server occasionally will delete a file at
random (that's actually what triggered the MSDB's own move from free.prohosting.com to paid hosting).
[Edited on 23-4-2004 by madscientist]
I weep at the sight of flaming acetic anhydride.
|
|
chemoleo
Biochemicus Energeticus
Posts: 3005
Registered: 23-7-2003
Location: England Germany
Member Is Offline
Mood: crystalline
|
|
the power of bioinformatics/alignments can NOT be understated!
Polverone, you have a point... if you consider madscience in a hands-on fashion. There is of course a lab connection from computational biology /
bioinformatics to the lab, but I fear it will be hard to implement that in an amateur lab. How would anyone here go about cloning a piece of DNA,
expressing proteins, studying interactions and so on? Without the proper funding (involving literally millions) it's impossible to do, just a
single Eppendorf pipette costs 80 pounds. Things such as nucleotides for cloning cost even more, and particularly the DNA polymerases. No way any
single person living on a student budget could afford this of course.
However... if you consider bioinformatics in a way that everything is done in silico, then I dont see a problem.... providing that a) the
person has a decent background knowledge on alignments etc are about b) that the person can be satisfied by finding new protein families and such, yet
not really being able to do anything with it, or publishing it and c) being totally abstract and confusing to anyone else here... and elsewhere! This
is why I struggle to explain anything I do in the lab to my parents, non-biochem friends etc...
Still, there is a VERY DEFINITE connection between bioinformatics and the lab. It's a shame you seem to get such an abstract course - it's
like math, it is by far not as interesting if it is pure math.. .but once applied, one can appreciate the power of it!
The power of DNA/protein sequence alignments / bioinformatics can not be underestimated! Consider this: I happen to identify this protein (this is
ficticious) which acts in the insulin regulation pathway (important for diabetes). I clone the DNA, and obtain the DNA sequence of this protein.
Running a quick database search for alignments with any other existing human proteins show that it has multiple domains, or functional subunits, that
occur in a wide range of completely unrelated proteins. Yet, because some of the domains were described in other proteins, I can already partially
predict how it works, and where it is likely to act!!!
Or consider this case, from a practical class that I did a few years back.
We identified a single transcription factor. Did sequence alignments with the whole human genome, plus all other genomes that are known to date.
This way we were able to pick out the regions in the protein sequence that are highly conserved (= little change), which indicates crucial
evolutionary importance (the more mutations, the less important this part of the sequence is). Then we constructed DNA primers BASED on the aligned
conserved regions, and tried to clone up ANY DNA sequence that has identical regions.
THis way we identified a huge transcription factor family, with >20 members! Each of those transcription factors of course warrant lots of
independent research... and explains our biochemistry increasingly better!
I hope that makes sense!
PS By the way, precisely for that reason I declined a PhD offer at the European Bioinformatics Institute (www.ebi.ac.uk), because I found out that NO actual experiments were done there! Purely computational biology! I decided too much computer work
(programming algorithms and such) should not be my future - I need to get my hands dirty !
Edit: Maybe this should go into the bioch. section after all ...
[Edited on 26-4-2004 by chemoleo]
Never Stop to Begin, and Never Begin to Stop...
Tolerance is good. But not with the intolerant! (Wilhelm Busch)
|
|
Polverone
Now celebrating 21 years of madness
Posts: 3186
Registered: 19-5-2002
Location: The Sunny Pacific Northwest
Member Is Offline
Mood: Waiting for spring
|
|
thanks for the explanation
It sounds like there really isn't much "poking" going on, in mad or professional science. Understanding and discovery are wonderful,
are necessary, are at the heart of science. But for Mad Science, after your months of hard work, you should have a monstrous Abomination Against
Nature to show for your hard work, as well as better understanding of biochemical pathways. It's okay if you haven't made any monsters yet;
we all need day jobs and publishable results.
I remember seeing pictures of genetically altered mice with human ears growing out of their backs some time ago. Now that's mad
science
PGP Key and corresponding e-mail address
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
Bioinformatics thread
This is an area which seems rich for home science, given the availability of cheap computers and an overabundance of molecular biology data.
Recently I have been exploring the BioPython module. (here: http://biopython.org/ ) This is a very nifty package which automates a lot of common computational processes for molecular biology, such as
sequence alignment, BLAST search, and PDB interfacing. I will note that some parts (eg, the Motif submodule) appear to be under active development
which is not necessarily reflected in the documentation/tutorial.
For directed learning, I recommend http://rosalind.info/ . This is a challenge-response style online programming course. It covers a lot of ground and gives a good overview of the
biochemical and computational background. My only complaint is that it can be a little sensitive to the format of answers (I am still trying to get it
to accept my answer to one problem) and that it doesn't let you answer certain questions out of order.
Any other cool bioinformatics resources/protips?
[Edited on 27-9-2013 by mayko]
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
Prior threads:
https://www.sciencemadness.org/whisper/viewthread.php?tid=19...
https://www.sciencemadness.org/whisper/viewthread.php?tid=75...
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
bfesser
|
Threads Merged 27-9-2013 at 15:11 |
Chemosynthesis
International Hazard
Posts: 1071
Registered: 26-9-2013
Member Is Offline
Mood: No Mood
|
|
Mayo (phone auto correcting your name) raises an excellent point of expense in tying bioinformatics with subsequent experimentation; much of my
bioinformatics experience has been in multidisciplinary groups involving molecular biology, pharmacology, and synthetic chemistry. Obviously, this
broad range of specialization requires more than one person to accomplish a project in any reasonable amount of time or grant schedule. I have
personally most used bioinformatics to identify homology models and potential drug targets for various pharmacophore screenings, which then requires
serious chemistry to discover or test QSAR for eventual commercial exploitation. Unfortunately, expense and zoning make much of the application
impossible on the amateur scale, but less ambitious projects (in silicon screening then proposed syntheses) are within the reach of a diligent and
educated amateur.
[Edited on 29-9-2013 by Chemosynthesis]
|
|
bfesser
|
Thread Moved 29-9-2013 at 07:07 |