|
|
Line 1: |
Line 1: |
- | [[Image:f3env.png|thumb]] (see [[Team:Paris/Modeling/Oscillations#Biochemical_Assumptions|the considerations on the use of EnvZ]])
| + | {{Paris/Menu}} |
| | | |
- | We have [EnvZ]<sub>produced</sub> = {coef<sub>env</sub>}''expr(pTet)'' = {coef<sub>env</sub>} ƒ1([aTc]<sub>i</sub>)
| + | {{Paris/Header|Method & Algorithm : ƒ6}} |
| | | |
- | and [EnvZ]<sub>total</sub> = [EnvZ]<sub>b</sub> + [EnvZ]<sub>produced</sub>
| + | [[Image:f6DCA.png|thumb|Specific Plasmid Characterisation for ƒ6]] |
| | | |
- | and [FliA] = {coef<sub>FliA</sub>}''expr(pBad)'' = {coef<sub>FliA</sub>} ƒ2([arab]<sub>i</sub>)
| + | We have <span style="color:#0000FF;">[''FlhDC'']<sub>''real''</sub> = {coef<sub>''flhDC''</sub>} ƒ1([aTc]<sub>i</sub>)</span> |
| + | and <span style="color:#0000FF;">[''FliA'']<sub>''real''</sub> = {coef<sub>''fliA''</sub>} ƒ2([arab]<sub>i</sub>)</span> |
| | | |
- | So, if we denote phosphorylated OmpR by ''OmpR<sup>*</sup>'', we have
| + | but we use <span style="color:#0000FF;">[aTc]<sub>i</sub> = Inv_ƒ1( [''FlhDC''] ) </span> |
| + | and <span style="color:#0000FF;">[arab]<sub>i</sub> = Inv_ƒ2( [''FliA''] ) </span> |
| | | |
- | [[Image:F3ompfromenv.jpg|center]]
| + | So, at steady-states, |
| | | |
- | that we can then introduce in the previous expression ([[Team:Paris/Modeling/f3|see ƒ3]]) :
| + | [[Image:F6.jpg|center]] |
| | | |
- | [[Image:F3ompfinalenv.jpg|center]]
| + | <br> |
| | | |
- | <br><br> | + | <div style="text-align: center"> |
| + | {{Paris/Toggle|Table|Team:Paris/Modeling/More_f6_Table}} |
| + | </div> |
| | | |
- | {|border="1" style="text-align: center"
| |
- | |param
| |
- | |signification
| |
- | |unit
| |
- | |value
| |
- | |comments
| |
- | |-
| |
- | |[expr(pFlhDC)]
| |
- | |expression rate of <br> pFlhDC '''with RBS E0032'''
| |
- | |nM.min<sup>-1</sup>
| |
- | |
| |
- | |need for 20 mesures with well choosen values of [aTc]<sub>i</sub>
| |
- | |-
| |
- | |γ<sub>GFP</sub>
| |
- | |dilution-degradation rate <br> of GFP(mut3b)
| |
- | |min<sup>-1</sup>
| |
- | |0.0198
| |
- | |
| |
- | |-
| |
- | |[GFP]
| |
- | |GFP concentration at steady-state
| |
- | |nM
| |
- | |
| |
- | |need for 20 mesures
| |
- | |-
| |
- | |(''fluorescence'')
| |
- | |value of the observed fluorescence
| |
- | |au
| |
- | |
| |
- | |need for 20 mesures
| |
- | |-
| |
- | |''conversion''
| |
- | |conversion ratio between <br> fluorescence and concentration
| |
- | |nM.au<sup>-1</sup>
| |
- | |(1/79.429)
| |
- | |
| |
- | |}
| |
| | | |
- | <br><br> | + | <div style="text-align: center"> |
| + | {{Paris/Toggle|Algorithm|Team:Paris/Modeling/More_FP_Algo}} |
| + | </div> |
| | | |
- | {|border="1" style="text-align: center"
| + | Then, if we have time, we want to verify the expected relation |
- | |param
| + | |
- | |signification <br> corresponding parameters in the [[Team:Paris/Modeling/Oscillations#Resulting_Equations|equations]]
| + | [[Image:SumpFlgA.jpg|center]] |
- | |unit
| + | |
- | |value
| + | <br> |
- | |comments
| + | |
- | |-
| + | <center> |
- | |K<sub>21</sub><sup>eff</sup>
| + | [[Team:Paris/Modeling/Implementation| <Back - to "Implementation" ]]| <br> |
- | |dissociation constant OmpR_-_EnvZ <br> K<sub>21</sub><sup>eff</sup>
| + | [[Team:Paris/Modeling/Protocol_Of_Characterization| <Back - to "Protocol Of Characterization" ]]| |
- | |nM
| + | </center> |
- | |
| + | |
- | |the literature [[Team:Paris/Modeling/Bibliography|[?] ]] gives K<sub>21</sub> =
| + | |
- | |-
| + | |
- | |n<sub>21</sub>
| + | |
- | |complexation order OmpR_-_EnvZ <br> n<sub>21</sub>
| + | |
- | |no dimension
| + | |
- | |
| + | |
- | |the literature [[Team:Paris/Modeling/Bibliography|[?] ]] gives n<sub>21</sub> =
| + | |
- | |-
| + | |
- | |{coef<sub>env</sub>}
| + | |
- | |coefficient due to the difference of the RBS and degradation rate between EnvZ and GFP <br> ! not precised in the equations !
| + | |
- | |no dimension
| + | |
- | |
| + | |
- | |! not precised in the equations ! watch out when writing the corresponding simulating program
| + | |
- | |-
| + | |
- | |[EnvZ]<sub>b</sub>
| + | |
- | |"basal" presence of EnvZ <br> [EnvZ]<sub>b</sub> | + | |
- | |nM
| + | |
- | |
| + | |
- | |the literature [[Team:Paris/Modeling/Bibliography|[?] ]] gives, under high osmolarity, [EnvZ]<sub>b</sub> =
| + | |
- | |-
| + | |
- | |[OmpR]<sub>b</sub>
| + | |
- | |"basal" presence of OmpR <br> [OmpR]<sub>b</sub>
| + | |
- | |nM | + | |
- | |
| + | |
- | |the literature [[Team:Paris/Modeling/Bibliography|[?] ]] gives, under high osmolarity, [OmpR]<sub>b</sub> =
| + | |
- | |}
| + | |
Method & Algorithm : 6
Specific Plasmid Characterisation for 6
We have [FlhDC]real = {coefflhDC} 1([aTc]i)
and [FliA]real = {coeffliA} 2([arab]i)
but we use [aTc]i = Inv_1( [FlhDC] )
and [arab]i = Inv_2( [FliA] )
So, at steady-states,
↓ Table ↑
param
| signification
| unit
| value
| comments
|
(fluorescence)
| value of the observed fluorescence
| au
|
| need for 20 mesures with well choosen values of [aTc]i and for 20 mesures with well choosen values of [arab]i and 5x5 measures for the relation below?
|
conversion
| conversion ratio between fluorescence and concentration ↓ gives ↓
| nM.au-1
| (1/79.429)
|
|
[GFP]
| GFP concentration at steady-state
| nM
|
|
|
γGFP
| dilution-degradation rate of GFP(mut3b) ↓ gives ↓
| min-1
| 0.0198
| Time Cell Division : 35 min.
|
6
| activity of pFlgA with RBS E0032
| nM.min-1
|
|
|
param
| signification corresponding parameters in the equations
| unit
| value
| comments
|
β26
| total transcription rate of FlhDC><pFlgA with RBS E0032 β26
| nM.min-1
|
|
|
(K3/{coeffliA})
| activation constant of FlhDC><pFliL K3
| nM
|
|
|
n3
| complexation order of FlhDC><pFliL n3
| no dimension
|
|
|
β27
| total transcription rate of FliA><pFliL with RBS E0032 β27
| nM.min-1
|
|
|
(K9/{coefflhDC})
| activation constant of FliA><pFliL K9
| nM
|
|
|
n9
| complexation order of FliA><pFliL n9
| no dimension
|
|
|
|
↓ Algorithm ↑
find_P
function optimal_parameters = find_FP(X_data, Y_data, initial_parameters)
function output = act_pProm(parameters, X_data)
for k = 1:length(X_data)
output(k) = parameters(1)*hill(X_data(k), parameters(2), parameters(3));
end
end
options=optimset('LevenbergMarquardt','on','TolX',1e-10,'MaxFunEvals',1e10,'TolFun',1e-10,'MaxIter',1e4);
optimal_parameters = lsqcurvefit( @(parameters, X_data) act_pProm(parameters, X_data),...
initial_parameters, X_data, Y_data, options );
end
|
Then, if we have time, we want to verify the expected relation
<Back - to "Implementation" |
<Back - to "Protocol Of Characterization" |
|