Team:Sheffield/Demonstrate
Demonstrate
To demonstrate that the device works, a bacterial growth experiment was performed using the constructed device to detect the changes in turbidity that occurred due to the bacterial growth. As the bacteria grow and the solution becomes more turbid, more light is attenuated as it passes through the sample in the well. The transmittance of light that reaches the photo-diode is therefore decreased and this decrease of light causes the photo-diode to produce less current across it. A decrease in the current across the photo-diode will cause the voltage across the photo-diode to also decrease. This change in voltage is measured by using a potential divider between the photo-diode and a resistor. As the voltage across the photo-diode decreases the voltage across the resistor increases, meaning if the bacteria are growing in the well, a voltage increase across the resistor will be seen. This is shown in the graph below:
Figure 1: The Figure above shows a bacterial growth curve that was measured with the constructed device. As bacteria grow and block the light the voltage across the photo-diode decreases and therefore the voltage across the resistor increases. The increase of the voltage across the resistor as the bacteria grow is what we can see above.
The slight unexpected variations are due to noise, however it can be removed by smoothing the data. To see more information about this please look at "Smoothing" under the software section. An example of a graph that has been smoothed is shown below:
Figure 2: The Figure above shows a bacterial growth after it has been smoothed to reduce noise.
To run the experiment above, different dilutions of bacteria were used. To do this a culture of E. coli was grown overnight. 1ml of this culture was then taken and diluted it in 9ml of LB media. This dilution went into all of the wells in the first row of a 96 well-plate. Then, 1ml of fluid was then taken from the first dilution and it was diluted again with another 9ml of LB media, and this dilution went into all the wells in the 2nd row of the plate. This process was repeated until the entire plate was filled. We did this so that we would have different concentrations of bacteria across the plate.
We repeated this process with 2 plates, using the same dilutions in each of the rows of the plate. We than ran a calibration experiment between our device, which contained 1 plate, and a Victor plate reader with the 2nd plate. As the 1st and 2nd plate were prepared the same, with the same bacteria culture the growth of the bacteria in the two plate would be the same, hence we are able to run a calibration experiment between the two devices with the two plates.
A calibration was performed by comparing some of the data retrieved from the constructed device to the data retrieved from the plate reader experiment. A graph of this calibration is shown below:
Figure 3: The graph above shows the data collected with a plate reader (on the y axis), and the corresponding data collected from the constructed device (on the x axis). The linear regression of this data is shown with the solid line, whereas the dotted lines so the 95% confidence intervals. This line has the equation: Y = 0.000665*X + 0.05062 and an R-squared value 0.9313
The calibration shows that there is a reasonably good fit for estimating the expected change for the absorbance read by the plate reader in accordance to the given change of voltage across the photo-diode from the constructed device. With an estimated R-squared value of 0.9313, this means you could use the constructed device to measure a bacterial growth curve with reasonable confidence that the bacteria growth curve measured with the plate reader is much the same.
