Scientific Computing

Discussion in 'MacBook Pro' started by farkasam, Aug 9, 2009.

  1. farkasam macrumors member

    Joined:
    Apr 10, 2009
    Location:
    New York
    #1
    Hi

    Does anyone here use their Mac for scientific apps, specifically genomics/proteomics applications. I am analysing microarray data and wanted to get a feel for what software/hardware set ups people might have experience with. Thanks :)
     
  2. libskate macrumors member

    Joined:
    Aug 9, 2009
    #2
    R

    I use my macbook pro for Statistical analysis of microarray data using the open-source R statistical software.
     
  3. Inside_line macrumors regular

    Joined:
    Jun 21, 2005
    Location:
    U.S.A.
    #3
    I've started using Ingenuity on it, but this is java based and also requires a subscription. I'd be interested to see what other responses you get on here... I would not suspect many 'omics people frequent these forums.
     
  4. admanimal macrumors 68040

    Joined:
    Apr 22, 2005
    #4
    Not sure what software you use, but I know that Matlab actually runs noticeably faster in Windows in a virtual machine than it does running natively in OS X on the same computer. Of course it generally runs OK in OS X too, but if speed is a concern then you are better off running it in Windows.
     
  5. Feng Shui macrumors regular

    Joined:
    Apr 9, 2008
    #5
    I've volunteered in a immunology lab. They're only equipped with Macs. Mind you, Macs are probably better than PC for scientific work.
     
  6. feelthefire macrumors 6502a

    Joined:
    Jun 13, 2006
    #6
    The microbiology lab I work in (my "real job") has 6 macs and 2 PCs. PCs control the equipment since most of that software isn't mac compatible, but we do everything else on MacBook Pros.
     
  7. sehnsucht77 macrumors 6502

    Joined:
    Dec 26, 2008
    #7
    i've used Macs and PCs extensively for research work and my macs routinely run Matlab for data analysis and my G4 and G5s run Labview through a National Instruments DAq rig. Overall, it is pretty stable but you need to make sure that your machines are spec'd properly for running the software - especially if you plan on integrating with hardware periphs. I also have some PCs running electrophys programs and Matlab as well. Other than my G5 running into some problems with a 3rd party program, all is well.

    @Feng Shui: I've used Macs and PCs for science work well over 10 years and I can say that one isn't necessarily better than the other for "all" tasks - you simply need to know what you are doing and you need to make sure the program is suited for the platform that you are using. Protein analysis, sequencing, data number crunching for electrophys..it is all the same and primarily dependent on your software because the computer basically processes what the software tells it to do. How well does the software package mesh with your machine? Is it thoroughly supported? I've built cell recording rigs from the ground up and these concerns are always a starting point for me before investing time and money into designing an effective research project. Moreover, Data analysis using stat software like Sigma Stat or SAS doesn't matter because those suites are available on Mac or PC so, again, as long as you know what you are doing and you are maintaining your machine properly, all should work well.

    @OP: is your microarray from ABI/ Invitrogen or from a company like Affymetrics? I've worked with both on Macs and PCs and the results do not matter as much on the type of machine as long as they are spec'd to handle the prog requirements. I had some samples run through an ABI array using an old Dell Latitude and the data was fine.
     
  8. farkasam thread starter macrumors member

    Joined:
    Apr 10, 2009
    Location:
    New York
    #8
    Nice input!

    Thanks much to all of you so far for the input. I agree that not many biomed scientists frequent these forums so it's nice to hear about the successes above.
     
  9. Akzel macrumors regular

    Joined:
    Mar 18, 2008
    #9
    We work with phylogeny and molecular evolution and our lab is more or less 50/50 Macs and Windows PCs. I use a Mac at the lab and at home and I have nothing to complain about it.

    The only complaint I've heard was from undergrads on the lack of a user friendly platform for alignments/sequence editing on the Mac platform (oh, the irony!), as most of them are taught using MEGA Software (Windows only, works like crap under Wine).

    Well, these days most of our analyses are performed on a central server, so in the end it's down to personal preferences which OS you use.
     
  10. Akzel macrumors regular

    Joined:
    Mar 18, 2008
    #10
    Forgot to say: For statistics we use Maxima, and it works fine either under Windows or OS X.
     
  11. farkasam thread starter macrumors member

    Joined:
    Apr 10, 2009
    Location:
    New York
    #11
    Ha! I dislike MEGA, mostly use Vector NCTI using Parallels for alignments. Or, have had good experience for basic tasks using web-based software (CLUSTALW and Boxshade). Thanks for the input.
     
  12. Akzel macrumors regular

    Joined:
    Mar 18, 2008
    #12
    Haha, I either use http://phylogeny.lirmm.fr/phylo_cgi/advanced.cgi for the quick stuff or I use clustalw/t-coffee "at home".

    I've avoid buying Parallels so far... :p
     
  13. sehnsucht77 macrumors 6502

    Joined:
    Dec 26, 2008
    #13
    i use vector NTI on my MBP and my PC at home. the progs work well for me.
     
  14. Schraiber macrumors member

    Joined:
    Jun 25, 2009
    #14
    I use matlab in parallels/windows boot. I understand that the mac native version is terrible :(.

    Some of my homies use R and they say it works great, though.
     
  15. m85476585 macrumors 65816

    Joined:
    Feb 26, 2008
    #15
    Have you tried the mac version? It really isn't *that* bad. The biggest issues I have run into are a slow GUI and a 2gb memory limit (the student version is 32 bit only for mac). Calculations might or might not be slower, but I haven't tested it.

    To get around the slow GUI, you can run matlab -nogui from the terminal and do everything in the terminal instead of the IDE. Doing that, it starts in about 10-15 seconds instead of a minute.
     

Share This Page