IGEM:IMPERIAL/2008/Prototype/Drylab: Difference between revisions
mNo edit summary |
|||
Line 23: | Line 23: | ||
Video images are captured into memory by the system at a basal video frame rate of 16.3Hz. This can be further increased by performing binning. | Video images are captured into memory by the system at a basal video frame rate of 16.3Hz. This can be further increased by performing binning. | ||
== Validating the tracking software == | |||
For our motility analysis we will be using ImageJ (open source freeware software written by NIH). We have considered a few tracking pluggins for ImageJ. The most accurate seems to be SpotTracker written by EPFL ( Ecole Polytechnique Federale de Lausanne), though not written specifically for cell tracking. Thus the need for us to validate the software and delimit its range of operability. From the video of the cells obtained via the wiedfield microscope, we determined using the measurement tool of ImageJ the average intensity and standard deviation of a few cells over a different frames. Identical measurements were made for the background. We identified a few static noisy patterns in the background as well. They might be due to optical effects. | |||
We generated a synthetic video consisting of a low intensity blob of changing shape, moving about a darker background. The SpotTracker failed to tracke the moving blob. Although it should have spotted the brightest spot first ( which is somewhere on the blob), it wrongly identifies point of coordinates (2,2) in pixels as being the brightest spot. We hypothesised that perhaps the blobs should not be of uniform intensity. | |||
=Data Analysis= | =Data Analysis= |
Revision as of 12:56, 21 August 2008
<html> <style type="text/css"> .firstHeading {display: none;} </style> </html> <html> <style type="text/css">
table.calendar { margin:0; padding:2px; }
table.calendar td { margin:0; padding:1px; vertical-align:top; } table.month .heading td { padding:1px; background-color:#FFFFFF; text-align:center; font-size:120%; font-weight:bold; } table.month .dow td { text-align:center; font-size:110%; } table.month td.today { background-color:#3366FF } table.month td {
border:2px; margin:0; padding:0pt 1.5pt; font-size:8pt; text-align:right; background-color:#FFFFFF; }
- bodyContent table.month a { background:none; padding:0 }
.day-active { font-weight:bold; } .day-empty { color:black; } </style> </html>
<html><a href=http://openwetware.org/wiki/IGEM:IMPERIAL/2008/Prototype><img width=50px src=http://openwetware.org/images/f/f2/Imperial_2008_Logo.png></img</a></html> | Home | The Project | B.subtilis Chassis | Wet Lab | Dry Lab | Notebook |
---|
<html> <style type="text/css"> div.Section { font:11pt/16pt Calibri, Verdana, Arial, Geneva, sans-serif; }
/* Text (paragraphs) */ div.Section p { font:11pt/16pt Calibri, Verdana, Arial, Geneva, sans-serif; text-align:justify; margin-top:0px; margin-left:30px; margin-right:30px; }
/* Headings */ div.Section h1 { font:22pt Calibri, Verdana, Arial, Geneva, sans-serif; text-align:left; color:#3366FF; }
/* Subheadings */ div.Section h2 { font:18pt Calibri, Verdana, Arial, Geneva, sans-serif; color:#3366FF; margin-left:5px; }
/* Subsubheadings */ div.Section h3 { font:16pt Calibri, Verdana, Arial, sans-serif; font-weight:bold; color:#3366FF; margin-left:10px; }
/* Subsubsubheadings */ div.Section h4 { font:12pt Calibri, Verdana, Arial, sans-serif; color:#3366FF; margin-left:15px; }
/* Subsubsubsubheadings */ div.Section h5 { font:12pt Calibri, Verdana, Arial, sans-serif; color:#3366FF; margin-left:20px; }
/* References */ div.Section h6 { font:12pt Calibri, Verdana, Arial, sans-serif; font-weight:bold; font-style:italic; color:#3366FF; margin-left:25px; }
/* Hyperlinks */ div.Section a {
}
div.Section a:hover {
}
/* Tables */ div.Section td { font:11pt/16pt Calibri, Verdana, Arial, Geneva, sans-serif; text-align:justify; vertical-align:top; padding:2px 4px 2px 4px; }
/* Lists */ div.Section li { font:11pt/16pt Calibri, Verdana, Arial, Geneva, sans-serif; text-align:left; margin-top:0px; margin-left:30px; margin-right:0px; }
/* TOC stuff */ table.toc { margin-left:10px; }
table.toc li { font: 11pt/16pt Calibri, Verdana, Arial, Geneva, sans-serif; text-align: justify; margin-top: 0px; margin-left:2px; margin-right:2px; }
/* [edit] links */ span.editsection { color:#BBBBBB; font-size:10pt; font-weight:normal; font-style:normal; vertical-align:bottom; } span.editsection a { color:#BBBBBB; font-size:10pt; font-weight:normal; font-style:normal; vertical-align:bottom; } span.editsection a:hover { color:#3366FF; font-size:10pt; font-weight:normal; font-style:normal; vertical-align:bottom; }
- sddm {
margin: 0; padding: 0; z-index: 30 }
- sddm li {
margin: 0; padding: 0; list-style: none; float: center; font: bold 12pt Calibri, Verdana, Arial, Geneva, sans-serif; border: 0px }
- sddm li a {
display: block; margin: 0px 0px 0px 0px; padding: 0 0 12px 0; background: #33bbff; color: #FFFFFF; text-align: center; text-decoration: none; }
- sddm li a:hover {
border: 0px }
- sddm div {
position: absolute; visibility: hidden; margin: 0; padding: 0; background: #33bbff; border: 1px solid #33bbff } #sddm div a { position: relative; display: block; margin: 0; padding: 5px 10px; width: auto; white-space: nowrap; text-align: left; text-decoration: none; background: #FFFFFF; color: #2875DE; font: 11pt Calibri, Verdana, Arial, Geneva, sans-serif } #sddm div a:hover { background: #33bbff; color: #FFFFFF } </style></html>
Team Strategy
We have divided the modelling team into 3 sections:
- Modelling Genetic Circuits - Erika
- Collecting Motility Data - Yanis
- Analysis of Motlity Data and Model Fitting - Clinton & Prudence
Modelling the genetic circuit
A simple ODE model was assumed in order to model the concentrations of the interacting proteins.
Motility Data Collection
We plan to capture video footage of B. subtilis and analyse its motility. Using a tracking software or algorithm, we would be able to determine the position of cells of interest at each frame. From this data, we can derive the run velocity and tumbling angle at each frame.
However, it is more complicated to derive the run time and tumbling time. If, for some frames the velocity is equal to zero, we may assume that this corresponds to the tumbling phase. Therefore, we can create a new array containing the tumbling duration for each frame. The same can be done for run time. To obtain run/tumbling duration over multiple frames, we have to find a sequence of frames that corresponds to the run phase or tumbling phase, and then sum their durations to determine the duration of the run/tumbling phases.
After obtaining data arrays with run/tumbling durations, run velocity and tumbling angle, we can then proceed on to data analysis.
Video Methods
We will be using the Zeiss Axiovert 200 inverted microscope with a fully motorised stage, controlled by Improvision Volocity acquisition software. This system offers a full incubation chamber with temperature and CO2 control, a large range of filter sets from UV to far-red and a highly sensitive 1300x1000 pixel camera for fast low-light imaging.
Video images are captured into memory by the system at a basal video frame rate of 16.3Hz. This can be further increased by performing binning.
Validating the tracking software
For our motility analysis we will be using ImageJ (open source freeware software written by NIH). We have considered a few tracking pluggins for ImageJ. The most accurate seems to be SpotTracker written by EPFL ( Ecole Polytechnique Federale de Lausanne), though not written specifically for cell tracking. Thus the need for us to validate the software and delimit its range of operability. From the video of the cells obtained via the wiedfield microscope, we determined using the measurement tool of ImageJ the average intensity and standard deviation of a few cells over a different frames. Identical measurements were made for the background. We identified a few static noisy patterns in the background as well. They might be due to optical effects.
We generated a synthetic video consisting of a low intensity blob of changing shape, moving about a darker background. The SpotTracker failed to tracke the moving blob. Although it should have spotted the brightest spot first ( which is somewhere on the blob), it wrongly identifies point of coordinates (2,2) in pixels as being the brightest spot. We hypothesised that perhaps the blobs should not be of uniform intensity.
Data Analysis
Our approach to modelling B.Subtilis motility is illustrated below:
- Gathering Data
The data which we will be interested in are: Run Velocity, Run Duration, Tumbling Angle and Tumbling Duration. These 4 parameters characterise the motility of bacteria, in this case B.Subtilis. We will collect data through the use of particle tracking software and self-generated / open source algorithms which convert raw x,y data coordinates into the above parameters of interest.
- Creating Alternative Models
Models of run velocity include a Gaussian Distribution, or the Maxwell Distribution which govern the velocities and energies of molecules. An exponential distribution may describe the memoryless characteristic of run duration. We will build up a database of models, for future model fitting.
- Fit Models to Data
In this first level of inference, we apply Bayes' Theorem. We first assume a particular model, and go on to derive the parameters of our model which maximises the data obtained.
- Assigning Preferences to Alternative Models
In this second level of inference, we use the evidence contributed by the data to compare fitted models. Using Occam's Razor, we are then able to deduce the best model which fits our data.
- Create New Models/Gather New Data
In the case where the models created do not fit the data effectively, we may consider gathering new data and creating new models. The process is repeated again if these steps are carried out.
- Future Actions
We will be characterising the B.Subtilis chasis, hence providing a valuable insight to the distributions and modelling of its motility.
Resources
The following are three tutorials which introduce us to data analysis and modelling. The tutorials are focused on the above approach.
Dry Lab Tutorial 1: Creating Synthetic Data
Dry Lab Tutorial 2: Statistical Data Analysis
Dry Lab Tutorial 3: Testing the Tracking Software
Dry Lab Tutorial 4: Modelling the Growth Curve
References
<html><center><table style="color:#ffffff;background-color:#66aadd;" cellpadding="3" cellspacing="1" border="0" bordercolor="#ffffff" align="center">
<tr><td><ul id="sddm"></html>[[IGEM:IMPERIAL/2008/New/{{{1}}}|< Previous]]<html></ul>
</td><td><ul id="sddm"><a href="#">Back to top</a></ul>
</td><td><ul id="sddm"></html>[[IGEM:IMPERIAL/2008/New/{{{2}}}|Next >]]<html></ul>
</td></tr></table>
</center></html>