Difference between revisions of "Team:William and Mary/Measurement"
| Line 28: | Line 28: | ||
<div style = 'padding-right: 14%; padding-left: 14%; text-indent: 50px;line-height: 25px;'> | <div style = 'padding-right: 14%; padding-left: 14%; text-indent: 50px;line-height: 25px;'> | ||
| − | This year team William and Mary's project focused on measuring the dynamical outputs of our decoding circuit. In the process we performed single cell time series measurements and <a href= 'https://2018.igem.org/Team:William_and_Mary/Chemical' style="color:green;">developed novel protocols for the removal of small molecule inducers</a>. Although single cell measurements using flow cytometry are our preferred way to obtain data, as our project developed, it became clear that much higher throughput would be required. As such, we spent a <a href = 'https://2018.igem.org/Team:William_and_Mary/Heat' style="color:green;">major portion</a> of our project <a href = 'https://2018.igem.org/Team:William_and_Mary/3G_Mixed' style="color:green;">creating methods</a> to test qualitative circuit behavior using a plate reader. Throughout the many design build test cycles our decoding system required, we have created many advances in measurement protocols that we hope will be useful to other teams working on complex and dynamic circuits. Additionally, we attempted to follow best practices for data availability and measurement by displaying our graphs in a manner that displays the underlying data, as well as providing full data (in .csv format) and experimental methods for every plot shown on the wiki. | + | This year team William and Mary's project focused on measuring the dynamical outputs of our decoding circuit. In the process we performed single cell time series measurements and <a href= 'https://2018.igem.org/Team:William_and_Mary/Chemical' style="color:green;">developed novel protocols for the removal of small molecule inducers</a>. Although single cell measurements using flow cytometry are our preferred way to obtain data, as our project developed, it became clear that much higher throughput would be required. As such, we spent a <a href = 'https://2018.igem.org/Team:William_and_Mary/Heat' style="color:green;">major portion</a> of our project <a href = 'https://2018.igem.org/Team:William_and_Mary/3G_Mixed' style="color:green;">creating methods</a> to test qualitative circuit behavior using a plate reader. Throughout the many design build test cycles our decoding system required, we have created many advances in measurement protocols that we hope will be useful to other teams working on complex and dynamic circuits. Additionally, we attempted to follow best practices for data availability and measurement by displaying our graphs in a manner that displays the underlying data, as well as providing <a href='https://2018.igem.org/Team:William_and_Mary/Experiments' style = "color: green;">full data (in .csv format) and experimental methods for every plot shown on the wiki</a>. |
</div> | </div> | ||
<div style='padding-top: 30px;'></div> | <div style='padding-top: 30px;'></div> | ||
| Line 60: | Line 60: | ||
<div style='padding-top: 0px;'></div> | <div style='padding-top: 0px;'></div> | ||
<div style = 'padding-right: 14%; padding-left: 14%; text-indent: 50px;line-height: 25px;'> | <div style = 'padding-right: 14%; padding-left: 14%; text-indent: 50px;line-height: 25px;'> | ||
| − | One crucial aspect of engineering disciplines is that data is displayed openly and in an unbiased fashion. To that end we displayed data in the appropriate fashion, using univariate scatter plots instead of bar graphs, and clearly displaying our data in a manner that allows the reader to say the level of variation present in our graphs [2]. As gene expression is log normally distributed [3], we made sure to use the geometric, rather than regular, means and standard deviations on our graphs, and made sure to use the appropriate y axis scale. Lastly, we attempt to provide a model for data availability by providing both the full data (in .csv or .fcs) format as well as detailed experimental methods for <u>every graph</u> on our wiki on a <a href=' | + | One crucial aspect of engineering disciplines is that data is displayed openly and in an unbiased fashion. To that end we displayed data in the appropriate fashion, using univariate scatter plots instead of bar graphs, and clearly displaying our data in a manner that allows the reader to say the level of variation present in our graphs [2]. As gene expression is log normally distributed [3], we made sure to use the geometric, rather than regular, means and standard deviations on our graphs, and made sure to use the appropriate y axis scale. Lastly, we attempt to provide a model for data availability by providing both the full data (in .csv or .fcs) format as well as detailed experimental methods for <u>every graph</u> on our wiki on a <a href='https://2018.igem.org/Team:William_and_Mary/Experiments' style = "color: green;">single page</a>.</div> |
<div style='padding-top: 30px;'></div> | <div style='padding-top: 30px;'></div> | ||
Revision as of 02:33, 17 October 2018
Measurement
Overview
This year team William and Mary's project focused on measuring the dynamical outputs of our decoding circuit. In the process we performed single cell time series measurements and developed novel protocols for the removal of small molecule inducers. Although single cell measurements using flow cytometry are our preferred way to obtain data, as our project developed, it became clear that much higher throughput would be required. As such, we spent a major portion of our project creating methods to test qualitative circuit behavior using a plate reader. Throughout the many design build test cycles our decoding system required, we have created many advances in measurement protocols that we hope will be useful to other teams working on complex and dynamic circuits. Additionally, we attempted to follow best practices for data availability and measurement by displaying our graphs in a manner that displays the underlying data, as well as providing full data (in .csv format) and experimental methods for every plot shown on the wiki.
Measurement Techniques
We contributed several novel measurement techniques to iGEM this year. First, we contributed to iGEM 3G Assembly, a new method of DNA assembly that will allow teams to easily create circuits with a wide variety of different parameters. This should not only serve to enhance future team's abilities to clone constructs, but should increase the ability of teams to characterize the properties of existing registry parts. Further, since this method is mostly compatible with CIDAR MoClo [1], we do not complicate characterization and standardization efforts by adding an additional and unnecessary standard. Indeed we demonstrated this potential by collaborating with Virginia iGEM to characterize two commonly used ribosome binding sites. Building off of this work, we also devised a way to utilize 3G to test different circuit design variants in a low cost, high throughput manner that requires no specialized equipment outside of a plate reader. Secondly, we believe that we have innovated in measurement by validating and using a protocol that allows for the ability to dynamically change the concentration of a small molecule inducers.
Lastly, we realized that many teams were unable to perform flow cytometry measurements due to lack of equipment or due to the prohibitive hourly costs associated with the use of their institutions flow cytometer. This issue is further exacerbated during time course measurements, as experiments often last multiple hours, which results in a large financial and time costs. To ameliorate this issue we designed tested and validated a flow cytometry protocol that preserves cells for later measurement without changing their fluorescence data (Figure 1). This procedure can be performed with nothing more than a -80C freezer and glycerol, enabling teams with limited resources to partner with more resourced teams to obtain single cell characterization of their projects, as well as enabling teams performing time course measurements to have a lower cost way to obtain their crucial single cell measurements.
Figure 1: Single cell fluorescence measurements of BBa_K2333428 time course live (NonFrozen) and 3 days later (Frozen). Each dot represents the geometric mean of at least 10,000 single cell measurements each of 3 biological replicates, the shaded region represents one geometric standard deviation above and below the geometric mean.
Data Visualization and Availability
One crucial aspect of engineering disciplines is that data is displayed openly and in an unbiased fashion. To that end we displayed data in the appropriate fashion, using univariate scatter plots instead of bar graphs, and clearly displaying our data in a manner that allows the reader to say the level of variation present in our graphs [2]. As gene expression is log normally distributed [3], we made sure to use the geometric, rather than regular, means and standard deviations on our graphs, and made sure to use the appropriate y axis scale. Lastly, we attempt to provide a model for data availability by providing both the full data (in .csv or .fcs) format as well as detailed experimental methods for every graph on our wiki on a single page.
References
[1] CIDAR MoClo: Improved MoClo Assembly Standard and New E. coli Part Library Enable Rapid Combinatorial Design for Synthetic and Traditional Biology. Sonya V. Iverson, Traci L. Haddock, Jacob Beal, and Douglas M. Densmore. ACS Synthetic Biology 2016 5 (1), 99-103 DOI: 10.1021/acssynbio.5b00124
[2] Weissgerber TL, Milic NM, Winham SJ, Garovic VD. Beyond bar and line graphs: time for a new data presentation paradigm. PLoS biology. 2015 Apr 22;13(4):e1002128.
[3] Beal J. Biochemical complexity drives log-normal variation in genetic expression. Engineering Biology. 2017 Jul 11;1(1):55-60.