UltraScan-II Instructional Videos

Video Topics:

Note: You can use these videos to teach yourself in a step-by-step fashion how to use UltraScan. The videos have been arranged roughly in a logical order, so it makes sense to watch them in the order they are presented. For more effective learning, the video player allows you to stop, pause, and rewind sections of the video in case it goes too fast for you.

Important: You will need audio on your computer to play these videos.

  1. Quickstart: Installation, UltraScan registration and US-LIMS database configuration

    This video shows how to install UltraScan, how to register UltraScan and install the registration file, and how to configure the UltraScan LIMS inside of UltraScan.

  2. UltraScan LIMS Account Creation

    The UltraScan software has 4 components:

    1. The UltraScan GUI software which is run on the desktop
    2. A web-based Laboratory Information Management System (LIMS)
    3. A database backend for data management and storage (remotely hosted by UTHSCSA)
    4. A supercomputer backend for demanding calculations (provided by UTHSCSA and NSF TeraGrid)

    The database backend allows you to associate data with an authenticated user, and this video shows how you can set up a password-protected account in your institutional database, which is used to manage your data and prevent others from accessing your data. If you do not yet find your institution listed here, please send an e-mail to Borries Demeler. The account is set up through the LIMS system.

  3. A few words about experimental noise

    Experimental noise degrades the quality of data, and consequently the reliability of the results. Therefore it is very important to minimize noise as much as possible. In addition to maintaining a properly functioning machine, UltraScan can remove and minimize certain types of noises to greatly improve the results obtained from the AUC. This video discusses the various kinds of noises and what can be done about them.

  4. Why you should measure velocity data in intensity mode and not in absorbance mode

    When using the UV-visible absorbance detector of the XLA you have the option of collecting absorbance or intensity data. This video discusses the significant and numerous benefits when you use intensity mode for the measurement of velocity data instead of absorbance mode.

  5. Conversion of Intensity Data to Pseudo-Absorbance Data

    After collection of data intensity mode (watch the previous video why you should use intensity mode for data collection), the data must be converted into pseudo-absorbance data before they can be processed further in UltraScan. This video shows you how you use UltraScan to perform this conversion on XLA data.

  6. Uploading of raw experimental data to the US-LIMS database

    After the data are converted from intensity to absorbance, the raw data need to be uploaded into the database. During the upload, a brief description and date is provided for each experiment, and each channel is associated with a database record for an analyte, a buffer composition and the investigator who owns the data. The buffer and analyte are used to provide automatic hydrodynamic corrections during analysis in UltraScan for density, viscosity and vbar.

  7. Editing velocity data using UltraScan

    After uploading of raw experimental data to the database, you need to retrieve the data back to a local analysis computer, and edit it before it can be used with the various analysis methods. In these videos, editing of absorbance, interference and fluorescence velocity data is demonstrated.

  8. Pre-processing velocity data with DCDT and uploading edited data to database

    After editing the data, we need to determine a sedimentation coefficient range to fit with the 2-dimensional spectrum analysis. Since intensity data contains a lot of time-invariant noise, DCDT is the ideal analysis method, since it eliminates time-invariant noise contributions from the data during analysis. After determining the s-value range, we need to upload the edited data to the database, so the supercomputer can access it.

  9. Submitting an initial 2DSA analysis from the US-LIMS web interface

    After uploading edited data, we are now ready to submit the first supercomputer analysis using the US-LIMS web interface. This video shows how to submit the analysis from the website and check the queue status.

  10. Retrieving 2DSA results and fitting the meniscus results and updating the meniscus

    The results from the initial 2DSA analysis returned by the supercomputer include

    1. an RMSD value for each fitted meniscus position
    2. a model file containing the time-invariant noise contributions
    3. a model file containing the finite element model that was fitted.
    This video shows how the RMSD values from the meniscus fit are used to obtain a meniscus position for each fitted cell, and how to retrieve the noise and model files from the e-mail.
  11. Reviewing 2DSA models and subtracting time-invariant noise

    After updating the meniscus position in our edited data, and retrieving both the finite element model and the time-invariant noise corrections from the initial 2DSA analysis, we can visualize the results and subtract time-invariant noise from the data.

  12. Refining the 2DSA analysis with a corrected meniscus

    Once the meniscus has been fitted and the initial time-invariant noise has been removed from the data, the noise/meniscus corrected data are uploaded again to the database to override the uncorrected data inside the database. At this point the analysis can be refined by repeating the time invariant noise correction in conjunction with the radial invariant noise correction. This video shows how to set up the refinement in the US-LIMS.

  13. Visualizing the refined solution (Refinement, part 2)

    The refined model and noise corrections are retrieved to the local harddrive and now visualized using the finite element model viewer. Time- and radially invariant noise vectors are subtracted again from the data and the data is re-uploaded once more to the database to replace the data with the refined version. This refinement represents the most optimal, noise-corrected solution that can be obtained for a given dataset and hence this solution should be used for all subsequent analyses.

  14. OPTIONAL: Performing the van Holde - Weischet analysis

    Once the data are cleaned up from all time- and radially invariant noise, and the meniscus position has been adjusted by fitting the meniscus, a van Holde - Weischet analysis can be performed to obtain a diffusion-corrected, model-independent sedimentation coefficient distribution.

  15. How to initialize the genetic algorithm with the 2-dimensional spectrum analysis results

    After obtaining a noise-corrected dataset, you can use the 2DSA results to initialize a genetic algorithm fit. The initialization values are saved to a file called "gadistro.dat", which can be uploaded to the US-LIMS for submission to the supercomputer.

  16. Submitting a genetic algorithm analysis with the US-LIMS

    Once the gadistro.dat file is defined, a genetic algorithm job is submitted using the US-LIMS. This is shown in the next video.

  17. Visualization of the genetic algorithm results

    Once the genetic algorithm analysis is completed, the resulting models are downloaded from the e-mail and the results are visualized and saved.

  18. Submitting a Monte Carlo analysis for the genetic algorithm and 2-dimensional spectrum analysis

    Results from the previous 2DSA and genetic algorithm analysis are still subject to random noise. The effect of this noise on the data can be quantified with a Monte Carlo analysis performed on top of the 2DSA or genetic algorithm analysis. Stochastic noise of the same quality as observed in the original experiment is added to the data and refitted. This process is repeated 20-50 times. Adding all models will amplify the intrinsic signal linearly, but stochastic noise only amplifies with a factor of square-root of 2. With each iteration, the Monte Carlo analysis hence amplifies the intrinsic signal when compared to the stochastic noise signal. Submission of a Monte Carlo analysis is shown in this video.

  19. Visualizing the 2DSA and GA Monte Carlo results

    The completed Monte Carlo analysis models are downloaded from the e-mail attachments and loaded into the finite element model viewer and saved.

Last Updated on ---- For questions please contact Borries Demeler