Tuesday, January 15, 2013
LEDs and Windows
I just received some LEDs I ordered for eBay and wanted to test them out because the last batched I ordered flashing LEDs by mistake. I didn't have a power source close by besides a microUSB plug connected to my Windows laptop so I connected the LED to the proper pins and it glowed. But then windows freaked out and wouldn't recognize any of my USB devices so I had to restart.
Let this be a lesson.
Let this be a lesson.
Cool Picture
http://fc09.deviantart.net/fs21/f/2007/285/e/6/e61c252f7534511d.jpg
Found it from here:
http://www.youtube.com/watch?feature=endscreen&v=2XG_0iV2B40&NR=1
Been on this big dubsteb kick lately while writing papers or coding.
Monday, January 14, 2013
Superman
Me trying to order something from Walgreens online. Yes, I shop as much as possible from home, don't hate.
Chat Transcript
Please wait while we find an agent to assist you...
Hi! My name is Khadija G. How may I help you?
Customer:
When I try to order online it keeps saying "The United States Postal
Service indicates your address has an incorrect apartment, suite, or
unit number. Please enter it below to continue."
Customer: no matter what I eneter
Khadija G:
I apologize for the inconvenience! Thank you for choosing
Walgreens.com. I will be more than happy to assist you. May I please
have the pleasure of knowing who I am chatting with today?
Customer: Does it matter?
Khadija G: Yes please I apologize but we do need to get your first and last name please.
Customer: Clark Kent
Khadija G: Thank you Clark are you entering the last 4 digits of your zip code?
Customer: You mean the extended zip code?
Customer: My zipcode is 60615
Customer: Well normally I reside in Metropolis New York and Work for the Daily Planet but currently I am in Chicago
Khadija G: Ok and where do you want the goods to be shipped too please?
Customer: ********** Apt 3L
Customer: Chicago, IL 60615
Customer: but you can't tell anyone my address
Customer: There is this guy Lex Luther who has been looking for me
Khadija G: Thank you Clark I promise not to tell anyone. Do you have a Walgreens.com account already?
Customer: No
Khadija G: Ok I would recommend opening a Walgreens.com account first this will allow you to enter both of your addresses.
Khadija G: I can send you the steps to register.
Customer: That's ok
Customer: I will do it myself
Customer: Thanks for the help
Khadija G: Thank you for chatting and please accept our apology for the inconvenience you experienced.
Customer: and keeping my identity safe
Khadija G: I understand Clark.
Customer: *wink* of course you do. Goodbye.
Khadija G: Goodbye Clark
Language Analysis
I have been working on a project for a while trying to analyze different modalities of the English language using Google Books, Twitter, IRC Chat and Phone conversations. The project is really quite interesting and has taught me a decent amount about statistics and linguistics. Lately I have been trying to create a confusion matrix with the different corpora in order to determine how different/similar these modalities of language we speak are. I am finally finishing up this project with what I consider respectable data and respectable results.
My goal now is to write up a nice little paper and send it off to some open access journal. I spent lots of yesterday coding and making some figures. I still love Perl despite its faults.
Sunday, January 13, 2013
Perl
Stupid Perl.
I think this is the second time I have encountered an error inherent in Perl.
This one has to do with the way Perl handles significant digits for arbitrarily long numbers. I was trying to make sure my confusion matrix program was running by comparing a language corpus against itself and it seemed to be a few numbers off in the matching. Finally I started comparing the actual values and like the 10th significant digit is off. Weird.
This is for my Language analysis project. I will write more on it soon. Now I need to think of a simple way to chop off 10 significant figures.
Hmmmm
I think this is the second time I have encountered an error inherent in Perl.
This one has to do with the way Perl handles significant digits for arbitrarily long numbers. I was trying to make sure my confusion matrix program was running by comparing a language corpus against itself and it seemed to be a few numbers off in the matching. Finally I started comparing the actual values and like the 10th significant digit is off. Weird.
This is for my Language analysis project. I will write more on it soon. Now I need to think of a simple way to chop off 10 significant figures.
Hmmmm
Friday, January 11, 2013
They trust me
BONUS! Well I guess there is one good thing about pretending to be smart.. people give you money.
I finally received the $2,500 for the Art/Science Fellowship/Fellows/Grant thing I received to work on building the Chromochord and the money is "no strings attached" I am super excited and now can buy lots of stuff.
I am working with a pretty cool composer, Francisco Castillo Trigueros.
Here is some of his music: https://www.youtube.com/watch?v=kZm0l4QDgto
I finally received the $2,500 for the Art/Science Fellowship/Fellows/Grant thing I received to work on building the Chromochord and the money is "no strings attached" I am super excited and now can buy lots of stuff.
I am working with a pretty cool composer, Francisco Castillo Trigueros.
Here is some of his music: https://www.youtube.com/watch?v=kZm0l4QDgto
TWINKIES!
I just thought they were joking when they said the Twinkies would be all gone. Isn't someone supposed to buy Hostess and continue to make Ding Dongs and Twinkies and Cup Cakes?
Anyways, I was given a box of Twinkies for my birthday which is coming up. I have been craving them for a bit. I could not find them anywhere! I can't believe the day has come where one cannot just walk into a gas station or grocer and buy some cup cakes or ding dongs.
Hopefully someone buys Hostess and continues to make the tasty treats.
Thursday, January 10, 2013
NMR
I forgot how much I like the Varian VNMRJ interface. We installed new Bruker consoles in our NMR facility and they are damn fancy there is like a bajillion buttons and stuff. I like VNMRJ because there is not a bajillion buttons in fact I think they opt against buttons.
One more HSQC and I have collected all the data for this paper! Well at least until the reviewers respond to us.
I pick up my check tomorrow for building a new version of the Chromochord. This means lots of cool science! I am excited and need to figure out what to buy. I was looking at the Nikon D3200 if I end up doing the Spectrophotography which I think at the moment is the best way to go.
Life is good when you have money to spend on cool science toys.
Going to do the elliptical today at the gym. I actually kind of enjoy it. I don't really enjoy running even though I play so much soccer.
Tuesday, January 8, 2013
My New Model of Protein Function
This is an email i sent out earlier. I thought it was interesting enough to post here. The residues and secondary structure elements are referenced from the protein I study Avena sativa LOV2.
----------------------------------
I think from our study and logically what appears to be a good model of protein function is one in which every residue is important in a context dependent manner. What this means is that one can mutate a residue but the protein can make up for it by certain residues having a new function or a new chi angle or something. If we mutate out one proposed pathway in our protein another can take over. Maybe the protein went through this transformation during evolution in the reverse. Maybe the beta sheets at one time were the only way to signal the Jalpha helix and then the A'a helix evolved to control the Jalpha because it is more energetically favourable than the beta sheets. The beta sheet function is still under slight selective pressure so it sticks around. Now if I disrupted the A'a helix the beta sheet function can take over. Imagine this on a smaller scale with residues. If I make L408A mutation and it removes the interaction between the A'a helix and Jalpha helix that is bad but maybe T407 forms a previously not found interaction due to the new phi/psi/chi angle it adopts and it appears from our mutation that L408A does nothing. The model is that every residue in a protein is context dependent but they all function in a thermodynamic way to give a protein it's structure and function. In one protein we might mutate a residue on the surface and nothing happens because other residues can easily make up for any loss of conformational energy or function. Maybe in another protein the residues can't. However, both residues could effectively have the same role. If I lose my pinky it would suck but my hand could compensate for it. If I lose my thumb it will definitely be a lot more difficult to make up for but both technically play a similar role.
----------------------------------
I think from our study and logically what appears to be a good model of protein function is one in which every residue is important in a context dependent manner. What this means is that one can mutate a residue but the protein can make up for it by certain residues having a new function or a new chi angle or something. If we mutate out one proposed pathway in our protein another can take over. Maybe the protein went through this transformation during evolution in the reverse. Maybe the beta sheets at one time were the only way to signal the Jalpha helix and then the A'a helix evolved to control the Jalpha because it is more energetically favourable than the beta sheets. The beta sheet function is still under slight selective pressure so it sticks around. Now if I disrupted the A'a helix the beta sheet function can take over. Imagine this on a smaller scale with residues. If I make L408A mutation and it removes the interaction between the A'a helix and Jalpha helix that is bad but maybe T407 forms a previously not found interaction due to the new phi/psi/chi angle it adopts and it appears from our mutation that L408A does nothing. The model is that every residue in a protein is context dependent but they all function in a thermodynamic way to give a protein it's structure and function. In one protein we might mutate a residue on the surface and nothing happens because other residues can easily make up for any loss of conformational energy or function. Maybe in another protein the residues can't. However, both residues could effectively have the same role. If I lose my pinky it would suck but my hand could compensate for it. If I lose my thumb it will definitely be a lot more difficult to make up for but both technically play a similar role.
Monday, January 7, 2013
Yeeaaaaa
http://www.youtube.com/watch?v=GrghtXWfVYM
I have a couple cool things to post on I just need to post the pictures and stuff!
Purifying my last protein for my paper today, woohoo! Only 107 constructs!
Sold a HeNe laser on eBay. I bought 27 of them for $107, crazy deal. Been playing around with them a little bit. Most I have tested are Red. No idea what to do with them yet or if I should just sell them all and make a bunch of money. I sold that one for $25. Yep, one of them for $25, hah.
Sending some a plasmid containing Taq polymerase to a guy in Estonia for DIYBio.
Have a meeting for the Art/Science Fellow thing tonight. Get to drink wine and act like a fool. I am pretty good at that. Glad we already have a prototype for our project! The Chromochord. BOOM!
Worked a little this weekend on the Spectrograph Camera I am really having a hard time wrapping my head around what wavelength of light determines a "color" or really if there even is any such thing as a "color". I didn't think this would be so much brain effort.
I have a couple cool things to post on I just need to post the pictures and stuff!
Purifying my last protein for my paper today, woohoo! Only 107 constructs!
Sold a HeNe laser on eBay. I bought 27 of them for $107, crazy deal. Been playing around with them a little bit. Most I have tested are Red. No idea what to do with them yet or if I should just sell them all and make a bunch of money. I sold that one for $25. Yep, one of them for $25, hah.
Sending some a plasmid containing Taq polymerase to a guy in Estonia for DIYBio.
Have a meeting for the Art/Science Fellow thing tonight. Get to drink wine and act like a fool. I am pretty good at that. Glad we already have a prototype for our project! The Chromochord. BOOM!
Worked a little this weekend on the Spectrograph Camera I am really having a hard time wrapping my head around what wavelength of light determines a "color" or really if there even is any such thing as a "color". I didn't think this would be so much brain effort.
Friday, January 4, 2013
Programming Languages
I was looking at the job requirements for some of the jobs at Dwave, the quantum computing company. For a software engineer they want someone that has experience with C and Python. Does that exist? I thought those things were mutually exclusive. Besides that fact they have the typically computer/hardware/software engineer requirements, i.e. we only want people skilled in every platform and every language in existence. haha, I thought people stopped asking for that.
Be able to write device drivers for Linux, check, Windows? WTF?
Experience with internet protocols, check, and RS485? WTF?
Don't worry I am not applying I just thought the requirements are kind of _out_ there.
Be able to write device drivers for Linux, check, Windows? WTF?
Experience with internet protocols, check, and RS485? WTF?
Don't worry I am not applying I just thought the requirements are kind of _out_ there.
Kinect
I spent like 6 hours last night trying to get my Kinect to work in Linux it seemed so easy from the OpenCV website. I should have known better. I actually got the Kinect video working the problem was with the body and skeleton tracking functions which are part of NiTE. When I downloaded the latest version of NiTE 2.0 there didn't seem to be any files in the download or rather the correct files! WTF! I didn't want to install an old version of NiTE for fear that it would break my installations of OpenNi. Dahhh!
We will see how much more effort I put into that.
I need to express my last protein today before I submit this paper we have been sitting on forever.
I fixed a Fluorescence plate reader the other day, it had a bad power supply need to figure out what to do with it.
I need to find a decent camera for my spectrography project and wait till my money arrives from this Fellowship/Grant thing. Until then I need to find another project.
We will see how much more effort I put into that.
I need to express my last protein today before I submit this paper we have been sitting on forever.
I fixed a Fluorescence plate reader the other day, it had a bad power supply need to figure out what to do with it.
I need to find a decent camera for my spectrography project and wait till my money arrives from this Fellowship/Grant thing. Until then I need to find another project.
Thursday, January 3, 2013
Hmmm
Was in early today. Just relaxing a little bit figuring out what I am going to do today.
Been looking for a nice low power mini PC that can run Linux that I can use for Robotics and I think the Beagleboard XM is the best thing out there at moment. I think the main problem is that I don't want to spend 87625171651276 hours just trying to install Linux so I can run OpenCV and bullshit. Raspberry pi is backordered like hell and in all reality is really slow and doesn't seem to be able to compete with the Beagleboard. The Beagleboard xM is however over $100 but you get what you pay for. It also has really nice Linux support and alot of developers. It has plenty of GPIO pins. I think it should work out great.
I think the current problem with technology when developing something is that lots of it needs to be done for scratch. yeah it would be great to use an Android phone but you basically have to port a whole operating system first. I probably spend most of my time searching for the proper tools to build what I want to build than in actually building it!
I used to like to build things from scratch like when I wrote IP Sorcery and refused to use libnet. It could have made things alot easier but I wanted more control and to learn how to code better. If I had to do it again I probably would have coded it the same way but nowadays I think alot more about how not to waste my time on stuff I can just spend $50 on.
Been looking for a nice low power mini PC that can run Linux that I can use for Robotics and I think the Beagleboard XM is the best thing out there at moment. I think the main problem is that I don't want to spend 87625171651276 hours just trying to install Linux so I can run OpenCV and bullshit. Raspberry pi is backordered like hell and in all reality is really slow and doesn't seem to be able to compete with the Beagleboard. The Beagleboard xM is however over $100 but you get what you pay for. It also has really nice Linux support and alot of developers. It has plenty of GPIO pins. I think it should work out great.
I think the current problem with technology when developing something is that lots of it needs to be done for scratch. yeah it would be great to use an Android phone but you basically have to port a whole operating system first. I probably spend most of my time searching for the proper tools to build what I want to build than in actually building it!
I used to like to build things from scratch like when I wrote IP Sorcery and refused to use libnet. It could have made things alot easier but I wanted more control and to learn how to code better. If I had to do it again I probably would have coded it the same way but nowadays I think alot more about how not to waste my time on stuff I can just spend $50 on.
Subscribe to:
Posts (Atom)



