diff --git a/content/12.model.md b/content/12.model.md
index eaf60a0ea869b75b02380207c809236f754416aa..bc2428aa3b278f22e25d2c130287598ee8a5e522 100644
--- a/content/12.model.md
+++ b/content/12.model.md
@@ -1,25 +1,25 @@
 <center>
 
-# **Hybrid Promoter Predicting Model**
+## **HPP Model**
 
 </center>
 
 <br>
 
 
-Our main goal is to filter out a promoter sequence with the best distinction under conditions with or without environmental oxygen, so that the survival of E.coli differs most in the tumor and the normal homeoenvironment. To achieve this goal, we divide the whole task into the following 3 steps:
+Our main goal is to filter out a promoter sequence with the best distinction under conditions with or without environmental oxygen, so that the survival of E.coli differs most in the tumor and the normal homeoenvironment. To achieve this goal, we develop <strong><font color=#8B0012>H</font></strong>ybrid <strong><font color=#8B0012>P</font></strong>romoter <strong><font color=#8B0012>P</font></strong>redicting (HPP) model. We divide the whole task into the following 3 steps:
 
 <br>
 
-##### **1. Predict the relative intensity of a set of pPept parts in promoters in aerobic environments**
+###### **1. Predict the relative intensity of a set of pPept parts in promoters in aerobic environments**
 
-##### **2. Predict the promoter intensity in hypoxia environment focusing primarily on the overall gene circuit, without taking into account the details such as mutant pPept components.**
+###### **2. Predict the promoter intensity in hypoxia environment focusing primarily on the overall gene circuit, without taking into account the details such as mutant pPept components.**
 
-##### **3. Predict the overall relative intensity by multiplying the 2 above**
+###### **3. Predict the overall relative intensity by multiplying the 2 above**
 
 <br>
 
-## **Aerobic Intensity Prediction** - Linear Model
+### **Aerobic Intensity Prediction** - Linear Model
 
 <br>
 
@@ -27,7 +27,7 @@ We calculate the transcription rate of the naked DNA strings, namely the intensi
 
 <br>
 
-### **Introduction**
+#### **Introduction**
 
 <br>
 
@@ -35,11 +35,11 @@ Inspiration from the paper [LaFleur, T.L., Hossain, A. & Salis, H.M. Automated m
 
 <br>
 
-### **Method**
+#### **Method**
 
 <br>
 
-#### **statistical thermodynamic model**
+##### **statistical thermodynamic model**
 
 <br>
 
@@ -93,7 +93,7 @@ Note that we determine the parts by choosing the -10 site as standard. Then, a f
 
  
 
-### **Result**
+#### **Result**
 
 | Promoter Type | Sequence_Name | dG_total | TX_rate | relative TX |
 | ------------- | :------------ | -------: | ------: | ----------: |
@@ -132,7 +132,7 @@ Since the culture environment varies with batch, which results in huge differenc
 
  <br>
 
-### **Discussion**
+#### **Discussion**
 
 <br>
 
@@ -145,30 +145,30 @@ Since the culture environment varies with batch, which results in huge differenc
  
 <br>
 
-## **And-Gate Modeling**
+### **And-Gate Model**
 
 <br>
 
-### **1.Model Construction**
+#### **1.Model Construction**
 
 We constructed the model based on the molecular mechanism of gene circuit of AND-gate, and gave the predicted transcription rate at arbitrarily given lactate and oxygen concentration. Thus, we theoretically demonstrated the selectivity and specificity of our engineered E.coli for tumor microenvironment. Also, results of modeling gave clues about methods for gene circuit optimization.
 
-#### **1.1 Landscape of the Circuit**
+##### **1.1 Landscape of the Circuit**
 
 <br>
 
 Gene circuit of our project consists of three main parts: 
 
 
-##### <strong>(1) Element targeting lack of oxygen:</strong>
+###### <strong>(1) Element targeting lack of oxygen:</strong>
 
 FNR which is an activator, binds specific sequence at the upstream site of pPepT, and helps RNA polymerase bindng onto the promoter. FNR protein is an activator when it exist as a dimer, and oxygen molecular hinders the polymerization of two FNR monomers.
 
-##### <strong>(2)Element targeting high concentration of lactate:</strong>
+###### <strong>(2)Element targeting high concentration of lactate:</strong>
 
 LldR which is a repressor, binds onto O1 and O2 sequence and subsequently polymerize into a dimer. The dimer forms a DNA loop and hinder the expression of genes between O1 and O2.
 
-##### <strong>(3)Element expressing LldR:</strong>
+###### <strong>(3)Element expressing LldR:</strong>
 
 With LldR overexpressing, pPepT is locked by DNA loop and will not be turned on.
 
@@ -178,12 +178,12 @@ Element (1) and (2) make up the and-gate structure. Promoter pPepT is turned on
   <img src="landscape.jpg" alt="Image 1" style="width: 85%; height: auto;">
 </div>
 
-##### Figure 1. Landscape of Engineered E.coli's Gene Circuit
+###### Figure 1. Landscape of Engineered E.coli's Gene Circuit
 
 <br>
 
 
-#### **1.2 Basic assumptions**
+##### **1.2 Basic assumptions**
 
 <br>
 
@@ -201,7 +201,7 @@ $$ Transcription \ rate = \sum{r_i}{p_i}$$
 
 <br>
 
-#### **1.3 System States and Corresponding Probability**
+##### **1.3 System States and Corresponding Probability**
 
 <br>
 
@@ -209,7 +209,7 @@ According to the gene circuits of our AND-gate, we abstracted 13 discrete states
 
 <br>
 
-##### Table1. Pictures of all States of the system and their corresponding probabilities.
+###### Table 3. Pictures of all States of the system and their corresponding probabilities.
 | State Number | State Picture | Statistical Weight | Variables |
 | --| -- | -- | -- |
 | 1| ![Img](/statepicture/state1.png) | 1 | $p_{none}$ |
@@ -227,20 +227,27 @@ According to the gene circuits of our AND-gate, we abstracted 13 discrete states
 | 13| ![Img](/statepicture/state13.png) | ${({{LldR} \over {N_{NS}}})^2}{e^{-{{2 \Delta e_{Rd}+{F_{loop}}+{\Delta e_{LL}}} \over {{k_b}T}}}}$ | $p_{ring}$ 
 
 Given statistical weights for all states, the probability of each state can be calculated by:
+
+<br>
+
+<center>
+
 $$ p_i = { {statistic \ weight_i} \over {\sum\limits_{j = 1}^{13} {statistic \ weight_j}} }$$
 
+</center>
+
 <br>
 
-#### **1.4 Parameter Determination**
+##### **1.4 Parameter Determination**
 
 <br>
 
-##### **Table2. All parameters and their values, units and references.**
+###### **Table 4. All parameters and their values, units and references.**
 
 
 | Parameter | Value | Unit | Significance |Reference |
 | --| -- | -- | -- | -- |
-| $N_{NS}$| $5 \times 10^6$ | - | Number of non-spoecific binding sites on the DNA strand. | 10.1016/j.gde.2005.02.006.  |
+| $N_{NS}$| 5e6 | - | Number of non-spoecific binding sites on the DNA strand. | 10.1016/j.gde.2005.02.006.  |
 | $RNAP$| $1000$ | - | Number of RNA polymerases | 10.1016/j.gde.2005.02.006.  |
 | $LldR$| $500$ | - | Number of LldR molecules| 10.1016/j.gde.2005.02.006.  |
 | $FNR$| $1000$ | - | Total number of FNR molecules, calculated by its monomers. | Manually setting |
@@ -259,11 +266,11 @@ $$ p_i = { {statistic \ weight_i} \over {\sum\limits_{j = 1}^{13} {statistic \ w
 <br>
 
 
-### **2. Effect of Oxygen**
+#### **2. Effect of Oxygen**
 
 <br>
 
-#### **2.1 Mechanism of FNR Dimer Formation**
+##### **2.1 Mechanism of FNR Dimer Formation**
 
 <br>
 
@@ -276,11 +283,11 @@ FNR activates RNA polymerase binding only when it exist as a dimer. In FNR dimer
 </div>
 
 
-##### Figure 2. Mechanism of depression of oxygen on FNR polymerization. 
+###### Figure 2. Mechanism of depression of oxygen on FNR polymerization. 
 
 <br>
 
-#### **2.2 Modeling of oxygen effect**
+##### **2.2 Modeling of oxygen effect**
 
 <br>
 
@@ -300,7 +307,7 @@ which means number of $FNR_{dim}$ and $FNR_{mon}$ can be settled by equilibrium
   <img src="/outputfigure/and-fig-fnrdim-function.png" alt="Image 1" style="width: 80%; height: auto;">
 </div>
 
-##### Figure 3. Left: Relation between $[FNR]_{dim}$ and $K_{FNR}$. Right: Relation between $[FNR]_{dim}$ and $oxygen$.
+###### Figure 3. Left: Relation between $[FNR]_{dim}$ and $K_{FNR}$. Right: Relation between $[FNR]_{dim}$ and $oxygen$.
 
 <br>
 
@@ -321,16 +328,16 @@ $$const_{oxy,a} = 50, \ const_{oxy,b} = 30, \ \ oxygen \in [0, \ 0.3]$$
 
 
 
-##### Figure 4. Left: Probabilities of the basic four states (none, FNR, RNAP, FNR_RNAP) under condition of different $[FNR]_{dim}$. Purple: None of proteins bounded onto DNA. Orange: RNAP bounded onto DNA. Yellow: FNR and RNAP bounded onto DNA together. Right: The probability distribution in the left figure changes as the number of FNR dimers increases. As the number of FNR dimers increases, the probability of RNAP binding state and FNR and RNAP co binding state increases, while the probability of no protein binding state decreases.
+###### Figure 4. Left: Probabilities of the basic four states (none, FNR, RNAP, FNR_RNAP) under condition of different $[FNR]_{dim}$. Purple: None of proteins bounded onto DNA. Orange: RNAP bounded onto DNA. Yellow: FNR and RNAP bounded onto DNA together. Right: The probability distribution in the left figure changes as the number of FNR dimers increases. As the number of FNR dimers increases, the probability of RNAP binding state and FNR and RNAP co binding state increases, while the probability of no protein binding state decreases.
 
 
 <br>
 
-### **3. Effect of Lactate**
+#### **3. Effect of Lactate**
 
 <br>
 
-#### **3.1 Mechanism of LldR Polymerization**
+##### **3.1 Mechanism of LldR Polymerization**
 
 <br>
 
@@ -343,84 +350,108 @@ The LldR protein monomer first forms a dimer, which can bind to O1 and O2. When
   <img src="/statepicture/lldr-lac.png" alt="Image 1" style="width: 50%; height: auto;">
 </div>
 
-##### Figure 5. Structure of LldR and the process of LldR polymerazation and dissociation.
+###### Figure 5. Structure of LldR and the process of LldR polymerazation and dissociation.
 
 
 <br>
 
-#### **3.2 Modeling of lactate effect**
+##### **3.2 Modeling of lactate effect**
 
 <br>
 
+<font style="font-size: 18px">
+
 $$ \Delta {e_{ll}^{'}} = \Delta e_{ll} + const_{lac} \times lactate $$
 $$ \Delta {e_{ld}^{'}} = \Delta e_{ld} + const_{lac} \times lactate$$
 $$const_{lac} = 0.3,\ \ \ lactate \in [0, \ 30]$$
-    
-<br>
-
-### **4. Result**
 
-<br>
-
-#### **4.1 Probability Distribution of All States as Lactate Concentration Changing**
+</font>
 
 <br>
 
-Applying our AND-gate model, we presented the probability of different states with fixed oxygen concentration of $0.1$ and lactate concentration ranging from 0.1mM, 1mM, 10mM, 30mM (Figure 5). We can see from the probability distribution diagram that when the concentration of lactic acid is low (0.1mM, 1mM), the system mainly exists in the form of DNA-loop. When the concentration of lactic acid is high (10mM, 30mM), the probability of DNA-loop appearing is almost zero, indicating good lactate concentration targeting. And under high lactate concentration conditions, the proportion of the four basic states (none, FNR, RNAP, FNR_RNAP) remains unchanged, and the probability of FNR and RNAP binding together is higher than the other three states. When the concentration of lactic acid increases to 30mM, the probability of FNR_RNAP state occurrence is much higher than other states, although when concentration of oxygen is not that low. In addition, when the lactate concentration is near the 10mM level, states of one LldR binding occurs, while when the lactate concentration is too low and too high, these states bear extreme low probabilities.
+#### **4. Effect of O1, O2 and pPepT Location**
 
 <br>
 
-<center>
-
-![Img](/outputfigure/and-figure-stateprob.png)
-
-</center>
-
-##### Figure 5. Probability distribution of all states at different conditions.
+In wet lab of our project, we constructed 6 kinds of composite promoters: $S3$, $S2$, $S1$, $NS$, $L2$, $L1$. These six composite promoters bears different base number between O1-pPepT and O1-O2 (Table3). 
 
 <br>
 
-#### **4.2 Transcription Rate Prediction**
+###### Table 5. Base number of O1-pPepT and O1-O2 for 6 constructed composite promoters.
+| Name of promoter | O1-pPepT (bp) | O1-O2 (bp) | $\lambda$ | $F_{loop}$
+| --- | --- | --- | --|--|
+| $S3$ | 50 | 134 | 0.373 | 32.45|
+| $S2$ | 61 | 143 | 0.427 | 39.84|
+| $S1$ | 52 | 143 | 0.364 | 28.95|
+| $NS$ | 71 | 162 | 0.438 | 37.01|
+| $L2$ | 62 | 153 | 0.405 | 33.50|
+| $L1$ | 81 | 172 | 0.471 | 40.31|
 
-<br>
+In the original model, parameter $F_{loop}$ bears the significance of the energy difference between states of DNA loop restricted by LldR and free DNA strand. With the distance of O1-pPepT and O1-O2 changed, energy difference between the two states changed. Considering both the length of composite promoter (base number between O1 and O2) and the location of pPepT (base number between O1 and pPepT) affect the configuration of DNA loop, we introduced a variable $\lambda$ to quantify this affect:
+$$ \lambda = {{x_{O1,pPepT}} \over {{x_{O1,O2}}}}$$
+$$ F_{loop} =  F_{loop, standard} \times {({{\lambda}/0.16})^2 \over {length}/160}, \ \ F_{loop,standard} = 5$$
 
-As the oxygen concentration decreases and the lactate concentration increases, the transcription rate of pPepT increases. The sudden jump in transcription rate accompanied by an increase in lactate concentration occurs around 7mM, while the sudden jump in transcription rate accompanied by an increase in oxygen concentration occurs around 0.18. It can be seen that the transcription rate has a greater sudden change in lactate concentration. (Figure 6)
+<div style="display: flex; justify-content: space-between;">
+  <img src="floop-o1o2-floop-lambda.png" alt="Image 1" style="width: 57%; height: auto;">
+  <img src="floop-surface.png" alt="Image 2" style="width: 43%; height: auto;">
+</div>
 
-<br>
+###### Figure 6. Left: Fixing $\lambda$, the functional relationship between $F_{loop}$ and base number between O1 and O2. Middle: Fixing base number between O1 and O2, the functional relationship between $F_{loop}$ and $\lambda$. Right: The surface of $F_{loop}$ with respect to changes in $\lambda$ and base number between O1 and O2.
 
-Considering that the oxygen concentration in normal tissues is about 0.2 and the lactate concentration is below 10mM, this result theoretically demonstrates the targeting of engineered Escherichia coli in low oxygen and high lactate environments corresponding to cancer.
+We changed the parameter $F_{loop}$ according to Table 3, and simulated the transcription rate distribution again (Figure 11). Under the influence of the length and pPepT position of the composite promoter, $F_{loop}$ changing result in different probabilities of circular DNA appearance, thereby affecting the sensitivity of the composite promoter to lactate. When the position of pPepT is closer to the position of O1, $\lambda$ is smaller, and $F_{loop}$ is smaller, and the circular DNA state appears as a larger probability. At these situations, the sudden changes in composite promoters' transcription rate caused by lactate concentration are more significant, as shown in the S3, S1, and L2 promoters in Figure 11. Compared to the initial value of $F_{loop}$, the order of magnitude of this parameter (about 30-40 $k_bT$) is slightly greater than the value of the original setting of $F_{loop}$ (about 3-5 $k_bT$). This change reduces the probability of the appearance of a circular DNA state, regardless of the oxygen and lactate concentrations. Compared with the initial parameter setting of $F_{loop}$, the current parameter value of $F_{loop}$ is in better agreement with the promoter characterization results in the wet experiment.
 
 <br>
 
-<center>
-
-![Img](/outputfigure/and-figure-heatmap.png)
-
-</center>
-
-##### Figure 6. Heatmap (left) and 3D (right)picture of predicted transcription rate of pPepT at all probable conditions, $oxygen \in [0, \ 0.3]$, $lactate \in [0 mM, 30 mM]$.
+<div style="display: flex; justify-content: space-between;">
+  <img src="s3-heat.png" alt="Image 1" style="width: 30%; height: auto;">
+  <img src="s2-heat.png" alt="Image 2" style="width: 30%; height: auto;">
+  <img src="s1-heat.png" alt="Image 4" style="width: 30%; height: auto;">
+</div>
 
-<br>
+<div style="display: flex; justify-content: space-between;">
+  <img src="s3-dis.png" alt="Image 1" style="width: 30%; height: auto;">
+  <img src="s2-dis.png" alt="Image 2" style="width: 30%; height: auto;">
+  <img src="s1-dis.png" alt="Image 4" style="width: 30%; height: auto;">
+</div>
 
-Next, we use this model to match the experimental data (Figure 7). Firstly, it is assumed that there is a simple linear relationship between od value and transcription rate. Considering that the lactate concentrations during the experiment were 0.1mM, 1mM, and 10mM, we took the logarithm of lactate concentration and created a numerical image of transcription rate lactate concentration (Figure 7, right). In areas with lactate concentrations below 10mM, both experimental data and predicted images exhibit approximately linear changes in transcription rate with respect to lactate concentration logarithms. However, in the region around 10mM, theoretical predictions indicate a sudden jump in transcription rate. And the experimental data does not match it.
+<div style="display: flex; justify-content: space-between;">
+  <img src="s3-lac.png" alt="Image 1" style="width: 15%; height: auto;">
+  <img src="s3-loglac.png" alt="Image 2" style="width: 15%; height: auto;">
+  <img src="s2-lac.png" alt="Image 4" style="width: 15%; height: auto;">
+  <img src="s2-loglac.png" alt="Image 1" style="width: 15%; height: auto;">
+  <img src="s1-lac.png" alt="Image 2" style="width: 15%; height: auto;">
+  <img src="s1-loglac.png" alt="Image 4" style="width: 15%; height: auto;">
+</div>
 
 <br>
 
 <div style="display: flex; justify-content: space-between;">
-  <img src="/outputfigure/and-figure-datapre.png" alt="Image 1" style="width: 50%; height: auto;">
-  <img src="/outputfigure/and-figure-datapre-log.png" alt="Image 2" style="width: 50%; height: auto;">
+  <img src="ns-heat.png" alt="Image 1" style="width: 30%; height: auto;">
+  <img src="l2-heat.png" alt="Image 2" style="width: 30%; height: auto;">
+  <img src="l1-heat.png" alt="Image 4" style="width: 30%; height: auto;">
 </div>
 
-##### Figure 7.Left: Relationship of transcription rate and lactate, normal oxygen condition is shown in blue and low oxygen condition is shown in orange. Right: Relationship of transcription rate and log(lactate).
+<div style="display: flex; justify-content: space-between;">
+  <img src="ns-dis.png" alt="Image 1" style="width: 30%; height: auto;">
+  <img src="l2-dis.png" alt="Image 2" style="width: 30%; height: auto;">
+  <img src="l1-dis.png" alt="Image 4" style="width: 30%; height: auto;">
+</div>
 
-<br>
+<div style="display: flex; justify-content: space-between;">
+  <img src="ns-lac.png" alt="Image 1" style="width: 15%; height: auto;">
+  <img src="ns-loglac.png" alt="Image 2" style="width: 15%; height: auto;">
+  <img src="l2-lac.png" alt="Image 4" style="width: 15%; height: auto;">
+  <img src="l2-loglac.png" alt="Image 1" style="width: 15%; height: auto;">
+  <img src="l1-lac.png" alt="Image 2" style="width: 15%; height: auto;">
+  <img src="l1-loglac.png" alt="Image 4" style="width: 15%; height: auto;">
+</div>
 
 
-### **5. Parameter Analysis**
+###### Figure 7. Transcription rate distribution of 6 different composite promoters, and the relationship between transcription rate and concentration of lactate, fixing oxygen concentration at 0 and 0.2. 
 
 <br>
 
-#### **5.1 Effect of $\Delta e_{FR}$**
+#### **5 Effect of $\Delta e_{FR}$**
 
 <br>
 
@@ -432,7 +463,7 @@ According to the distance between the FNR protein binding site and the RNA polym
   <img src="/statepicture/fr.png" alt="Image 1" style="width: 70%; height: auto;">
 </div>
 
-##### Figure 8. Molecular Mechanism of the Interaction between FNR Protein and RNA Polymerase
+###### Figure 8. Molecular Mechanism of the Interaction between FNR Protein and RNA Polymerase
 
 <br>
 
@@ -477,88 +508,77 @@ Based on this, we fixed other parameters in the model and adjust he value of $\D
 </div>
 
 
-##### Figure 9. Distribution of transcription rate with $\Delta e_{FR}$ takes the value of -6, -7, -9, -10, -11, -12 ${k_b}T$ units.
+###### Figure 9. Distribution of transcription rate with $\Delta e_{FR}$ takes the value of -6, -7, -9, -10, -11, -12 ${k_b}T$ units.
+
+<br>
 
 <br>
 
-#### **5.2 Effect of $F_{loop}$**
+#### **6. Result**
 
 <br>
 
-In wet lab of our project, we constructed 6 kinds of composite promoters: $s3$, $s2$, $s1$, $ns$, $l2$, $l1$. These six composite promoters bears different base number between O1-pPepT and O1-O2 (Table3). 
+##### **6.1 Probability Distribution of All States as Lactate Concentration Changing**
 
 <br>
 
-##### Table 3. Base number of O1-pPepT and O1-O2 for 6 constructed composite promoters.
-| Name of promoter | O1-pPepT (bp) | O1-O2 (bp) | $\lambda$ | $F_{loop}$
-| --- | --- | --- | --|--|
-| $s3$ | 50 | 134 | 0.530 | 39.750|
-| $s2$ | 61 | 143 | 0.573| 42.975|
-| $s1$ | 52 | 143 | 0.510| 38.250|
-| $ns$ | 71 | 162 | 0.568| 42.600|
-| $l2$ | 62 | 153 | 0.542| 40.650|
-| $l1$ | 81 | 172 | 0.593| 44.475|
+Applying our AND-gate model, we presented the probability of different states with fixed oxygen concentration of $0.1$ and lactate concentration ranging from 0.1mM, 1mM, 10mM, 30mM (Figure 5). We can see from the probability distribution diagram that when the concentration of lactic acid is low (0.1mM, 1mM), the system mainly exists in the form of DNA-loop. When the concentration of lactic acid is high (10mM, 30mM), the probability of DNA-loop appearing is almost zero, indicating good lactate concentration targeting. And under high lactate concentration conditions, the proportion of the four basic states (none, FNR, RNAP, FNR_RNAP) remains unchanged, and the probability of FNR and RNAP binding together is higher than the other three states. When the concentration of lactic acid increases to 30mM, the probability of FNR_RNAP state occurrence is much higher than other states, although when concentration of oxygen is not that low. In addition, when the lactate concentration is near the 10mM level, states of one LldR binding occurs, while when the lactate concentration is too low and too high, these states bear extreme low probabilities.
 
-In the original model, parameter $F_{loop}$ bears the significance of the energy difference between states of DNA loop restricted by LldR and free DNA strand. With the distance of O1-pPepT and O1-O2 changed, energy difference between the two states changed. Considering both the length of composite promoter and the location of pPepT affect the configuration of DNA loop, we introduced a variable $\lambda$ to quantify this affect:
-$$ \lambda = {{x_{O1,pPepT}+21bp} \over {{x_{O1,O2}}}+20bp}$$
-$$ F_{loop,fix} = {const_{F}} \times F_{loop} \times \lambda (1- \lambda), \ \ const_{F} = 25$$
+<br>
+
+<center>
+
+![Img](/outputfigure/and-figure-stateprob.png)
+
+</center>
 
-We fixed the parameter $F_{loop}$ according to Table 3, and simulated the transcription rate distribution again (Figure 10). Distribution of transcription rate does not change a lot with different $F_{loop}$ of different composite promoter.
+###### Figure 10. Probability distribution of all states at different conditions.
 
 <br>
 
-<div style="display: flex; justify-content: space-between;">
-  <img src="/site-figure/s3-heat.jpg" alt="Image 1" style="width: 33%; height: auto;">
-  <img src="/site-figure/s2-heat.jpg" alt="Image 2" style="width: 33%; height: auto;">
-  <img src="/site-figure/s1-heat.jpg" alt="Image 4" style="width: 33%; height: auto;">
-</div>
+##### **6.2 Transcription Rate Prediction**
 
-<div style="display: flex; justify-content: space-between;">
-  <img src="/site-figure/s3-od.jpg" alt="Image 1" style="width: 16%; height: auto;">
-  <img src="/site-figure/s3-transcription.jpg" alt="Image 1" style="width: 16%; height: auto;">
-  <img src="/site-figure/s2-od.jpg" alt="Image 2" style="width: 16%; height: auto;">
-  <img src="/site-figure/s2-transcription.jpg" alt="Image 2" style="width: 16%; height: auto;">
-  <img src="/site-figure/s1-od.jpg" alt="Image 4" style="width: 16%; height: auto;">
-  <img src="/site-figure/s1-transcription.jpg" alt="Image 4" style="width: 16%; height: auto;">
-</div>
+<br>
 
-<div style="display: flex; justify-content: space-between;">
-  <img src="/site-figure/ns-heat.jpg" alt="Image 1" style="width: 33%; height: auto;">
-  <img src="/site-figure/l1-heat.jpg" alt="Image 2" style="width: 33%; height: auto;">
-  <img src="/site-figure/l2-heat.jpg" alt="Image 4" style="width: 33%; height: auto;">
-</div>
+As the oxygen concentration decreases and the lactate concentration increases, the transcription rate of pPepT increases. The sudden jump in transcription rate accompanied by an increase in lactate concentration occurs around 7mM, while the sudden jump in transcription rate accompanied by an increase in oxygen concentration occurs around 0.18. It can be seen that the transcription rate has a greater sudden change in lactate concentration. (Figure 6)
 
-<div style="display: flex; justify-content: space-between;">
-  <img src="/site-figure/ns-od.jpg" alt="Image 1" style="width: 16%; height: auto;">
-  <img src="/site-figure/ns-transcription.jpg" alt="Image 1" style="width: 16%; height: auto;">
-  <img src="/site-figure/l1-od.jpg" alt="Image 2" style="width: 16%; height: auto;">
-  <img src="/site-figure/l1-transcription.jpg" alt="Image 2" style="width: 16%; height: auto;">
-  <img src="/site-figure/l2-od.jpg" alt="Image 4" style="width: 16%; height: auto;">
-  <img src="/site-figure/l2-transcription.jpg" alt="Image 4" style="width: 16%; height: auto;">
-</div>
+<br>
+
+Considering that the oxygen concentration in normal tissues is about 0.2 and the lactate concentration is below 10mM, this result theoretically demonstrates the targeting of engineered Escherichia coli in low oxygen and high lactate environments corresponding to cancer.
+
+<br>
+
+<center>
 
-##### Figure 10. Transcription rate distribution of 6 different composite promoters, and the relationship between Transcription rate and concentration of lactate, fixing oxygen concentration at 0 and 0.2. 
+![Img](/outputfigure/and-figure-heatmap.png)
+
+</center>
+
+###### Figure 11. Heatmap (left) and 3D (right)picture of predicted transcription rate of pPepT at all probable conditions, $oxygen \in [0, \ 0.3]$, $lactate \in [0 mM, 30 mM]$.
 
 <br>
 
-The simulation of the composite promoter length and pPepT promoter position using the AND gate model is highly consistent with real experimental data.
+Next, we use this model to match the experimental data (Figure 7). Firstly, it is assumed that there is a simple linear relationship between od value and transcription rate. Considering that the lactate concentrations during the experiment were 0.1mM, 1mM, and 10mM, we took the logarithm of lactate concentration and created a numerical image of transcription rate lactate concentration (Figure 7, right). In areas with lactate concentrations below 10mM, both experimental data and predicted images exhibit approximately linear changes in transcription rate with respect to lactate concentration logarithms. However, in the region around 10mM, theoretical predictions indicate a sudden jump in transcription rate. And the experimental data does not match it.
 
 <br>
 
 <div style="display: flex; justify-content: space-between;">
-  <img src="/site-figure/log-theory-site.jpg" alt="Image 1" style="width: 50%; height: auto;">
-  <img src="/site-figure/log-raw-site.png" alt="Image 2" style="width: 50%; height: auto;">
+  <img src="/outputfigure/and-figure-datapre.png" alt="Image 1" style="width: 50%; height: auto;">
+  <img src="/outputfigure/and-figure-datapre-log.png" alt="Image 2" style="width: 50%; height: auto;">
 </div>
 
-##### Figure 11. Left: Transcription rate of 6 kinds of composite promoters simulated by AND-gate model. Right: od-600 of 6 kinds of composite promoters (s3BFT, s2BFT, s1BFT, nsBFT, l2BFT, l1BFT) measured in wet lab. 
+###### Figure 12.Left: Relationship of transcription rate and lactate, normal oxygen condition is shown in blue and low oxygen condition is shown in orange. Right: Relationship of transcription rate and log(lactate).
 
 <br>
 
-### **6. Discussion**
+
+
+
+#### **7. Discussion**
 
 <br>
 
-#### **1. Comparison between wet lab and dry lab results.**
+##### **7.1 Comparison between wet lab and dry lab results.**
 
 <br>
 
@@ -566,7 +586,7 @@ The fitting results of the dry experiment indicate that regardless of how the bi
 
 <br>
 
-#### **2. Noise.**
+##### **7.2 Noise.**
 
 <br>
 
@@ -574,69 +594,91 @@ Biological processes, including gene transcription and expression processes, can
 
 <br>
 
-#### **3.Parameters' Physical Meaning.**
+##### **7.3 Parameters' Physical Meaning.**
 
 <br>
 
-When considering the interaction function of oxygen and lactate small molecule concentration, we referred to the specific molecular mechanisms of two small molecules participating in the pathway and proposed a conjecture of the interaction function based on the molecular mechanisms. However, due to the lack of direct measurement of the interactions between various proteins and DNA, we are unable to accurately calibrate const_ {ox, a}, const_ {ox, b} and const_ The specific values of the three parameters {lac} under different conditions do not currently have specific physical meanings, although they are physical quantities with certain units. This is the part where the model itself can be improved.
+When considering the interaction function of oxygen and lactate small molecule concentration, we referred to the specific molecular mechanisms of two small molecules participating in the pathway and proposed a conjecture of the interaction function based on the molecular mechanisms. However, due to the lack of direct measurement of the interactions between various proteins and DNA, we are unable to accurately calibrate $const_{oxy, a}$, $const_{oxy, b}$ and $const_{lac}$ The specific values of the three parameters lac under different conditions do not currently have specific physical meanings, although they are physical quantities with certain units. This is the part where the model itself can be improved.
 
 <br>
 
+<br>
 
-### **References**
+### Integrate Model
 
-<br>
+Aerobic intensity prediction model (step 1) provides relative transcription rates for different pPepT sequences, while AND-Gate model (step 2) provides relative transcription rates for different oxygen concentrations, lactate concentrations, and O1, pPepT, and O2 relative positions. For our initial goal of finding the most suitable composite promoter for targeting hypoxic and high lactate environments, we integrated the two models and presented the transcriptional rates of different promoters with different pPepT sequences as a function of oxygen and lactate conditions.
 
-1. Gao Y, Zhou H, Liu G, Wu J, Yuan Y, Shang A. Tumor Microenvironment: Lactic Acid Promotes Tumor Development. J Immunol Res. 2022 Jun 12;2022:3119375. doi: 10.1155/2022/3119375. PMID: 35733921; PMCID: PMC9207018.      
+6 promoters with different relative positions of O1, O2, and pPepT (s3, s2, s1, ns, l2, l1) were nested with 8 possible pPepT sequences, and 48 composite promoters were obtained by permutation and combination.  We assume that there is no mutual influence between the sequence targeting pPepT in the first step and the predictions targeting conditions, promoter length, and pPepT position in the second step. According to the multiplication principle of probability, the final prediction of hybrid promoter can be calculated by:
+$$Rate (oxygen, lactate, x_{O1,O2}, x_{O1,pPepT}, \alpha) = Rate_1(\alpha) \times  Rate_2(oxygen, lactate, x_{O1,O2}, x_{O1,pPepT})  $$
 
-2.  Brizel D. M., Schroeder T., Scher R. L., et al. Elevated tumor lactate concentrations predict for an increased risk of metastases in head-and-neck cancer. International Journal of Radiation Oncology • Biology • Physics . 2001;51(2):349–353. doi: 10.1016/S0360-3016(01)01630-3.
+###### Table 6. Predicted $Rate_2(\alpha)$ by step 1.
+| $\alpha$ | WA | WA | WT | WT | FA | FA | FT | FT |
+| -- | -- | -- | -- | -- | -- | -- | -- | -- |
+| $Name \ of \ pPepT$ | BWW | JWW | BWT | JWT | BFW | JFW | BFT | JFT |
+| $Rate_2(\alpha)$ | 388.241 | 379.012 | 980.648 | 957.338 | 724.798 | 707.569 | 1830.75 | 1787.23 |
 
-3. Bintu L, Buchler NE, Garcia HG, Gerland U, Hwa T, Kondev J, Phillips R. Transcriptional regulation by the numbers: models. Curr Opin Genet Dev. 2005 Apr;15(2):116-24. doi: 10.1016/j.gde.2005.02.007. PMID: 15797194; PMCID: PMC3482385.
+Given the two base number, lactate and oxygen conditions, $Rate_1$ can be calculated based on the AND-Gate model, and then give the pPepT sequence to obtain $Rate_2$. Multiplying the two rates, we obtain the predicted results of the integrated model.
 
-4. Mettert EL, Kiley PJ. Reassessing the Structure and Function Relationship of the O2 Sensing Transcription Factor FNR. Antioxid Redox Signal. 2018 Dec 20;29(18):1830-1840. doi: 10.1089/ars.2017.7365. Epub 2017 Nov 14. PMID: 28990402; PMCID: PMC6217745
+The results (Figure 12) indicate that all probable promoters have selectivity for lactate and oxygen conditions, but the basal expression levels of different promoters are different. S2, NS, and L1 bear higher basal expression levels. The basal expression levels of different promoters can be seen from the relative expression rate under low lactate concentration and normoxic conditions.
 
-5. Morrison M, Razo-Mejia M, Phillips R. Reconciling kinetic and thermodynamic models of bacterial transcription. PLoS Comput Biol. 2021 Jan 19;17(1):e1008572. doi: 10.1371/journal.pcbi.1008572. PMID: 33465069; PMCID: PMC7845990.
+In addition, regardless of the form of the composite promoter, the relative transcription rate ratio between different pPepT sequences is certain. This is determined by the model itself, as we assume that the composite promoter has no effect on the prediction of the pPepT sequence, i.e. the two rate variables $Rate_1$ and $Rate_2$ are independent of each other.
 
-6. Muller-Hill, B., 1996. The Lac Operon, first ed. de Gruyter, Berlin.
-Narang, A., Pilyugin, S.S., 2006. Why does the lac exhibit no bistability
-during growth of Escherichia coli on lactose or lactose + glucose? Bull.
-Math. Biol., submitted for publication.
+<div style="display: flex; justify-content: space-between;">
+  <img src="rate-all.png" alt="Image 1" style="width: 90%; height: auto;">
+</div>
+
+###### Figure 13. The relative transcription rates predicted by the integrated model for 48 possible promoters under hypoxic/aerobic conditions and different lactate concentrations.
+
+We then selected the composite promoters that were successfully characterized in the wet experiment and standardized the data in the same way as wet experiments. Due to the high background expression intensity of the S2, NS, and L1 promoters, after standardization, these three promoters showed lower selectivity for hypoxic and high lactate environments. The other three promoters S3, S1, and L2 have higher selectivity for hypoxic and high lactate environments. The transcription rates of different pPepT promoters maintain the proportion predicted by step1. After standardization, the differences in transcription rates among different pPepT sequences were eliminated. Because the relative transcription rate of each pPepT sequence under different conditions is standardized using data corresponding to its own sequence.
+
+<div style="display: flex; justify-content: space-between;">
+  <img src="rate-selected.png" alt="Image 1" style="width: 50%; height: auto;">
+  <img src="rate-selected-standardized.png" alt="Image 1" style="width: 50%; height: auto;">
+</div>
+
+###### Figure 14. Left: The relative transcription rates of all promoters successfully characterized in wet experiments under different conditions. Right: Standardized presentation of the data in the left image. The standardization method is to divide each row of data by the first column of that row. 
 
-7. Jensen D, Galburt EA. The Context-Dependent Influence of Promoter Sequence Motifs on Transcription Initiation Kinetics and Regulation. J Bacteriol. 2021 Mar 23;203(8):e00512-20. doi: 10.1128/JB.00512-20. PMID: 33139481; PMCID: PMC8088511.
+The high selectivity of S3, S1, and L2 towards high lactate and low oxygen environments is consistent with actual experimental results. 
 
-8. Wang J, Zhang S, Lu H, Xu H. Differential regulation of alternative promoters emerges from unified kinetics of enhancer-promoter interaction. Nat Commun. 2022 May 17;13(1):2714. doi: 10.1038/s41467-022-30315-6. PMID: 35581264; PMCID: PMC9114328.
+<div style="display: flex; justify-content: space-between;">
+  <img src="wet2.png" alt="Image 1" style="width: 45%; height: auto;">
+  <img src="rate-selected-standardized.png" alt="Image 1" style="width: 55%; height: auto;">
+</div>
 
-9. Oehler S, Amouyal M, Kolkhof P, von Wilcken Bergmann B, Hill BM. Quality and position of the three lac operators of E. coli define efficiency of repression. The EMBO journal. 1994; 13(14):3348–3355. https://doi.org/10.1002/j.1460-2075.1994.tb06637.x PMID: 8045263
+###### Figure 15. Experimental results (left) and predicted results (right) of standardized relative transcription rate.
 
-10. Oehler S, Eismann ER, Kra¨mer H, Hill BM. The three operators of the lac operon cooperate in repression. The EMBO journal. 1990; 9(4):973–979. https://doi.org/10.1002/j.1460-2075.1990.tb08199.x PMID: 2182324
 
-11. Bintu L, Buchler NE, Garcia HG, Gerland U, Hwa T, Kondev J, Kuhlman T, Phillips R. Transcriptional regulation by the numbers: applications. Curr Opin Genet Dev. 2005 Apr;15(2):125-35. doi: 10.1016/j.gde.2005.02.006. PMID: 15797195; PMCID: PMC3462814.
+The simulation results indicate that among the many successfully characterized promoters, L2-WT, S3-WT, and S3-FA are more in line with the model's predicted target for high lactate and low oxygen environments, indicating that the construction of these sequences is more successful.
+
 
 
 
 <br/>
 
 
-### **Code**
+#### **Code**
 
 <br>
 
 ::Pdf2
 ::
 
+<br>
 
-
-
+<center>
 
 ## **MC Model**
 
+</center>
+
 <br>
 
 ### **Introduction**
 
 <br>
 
-Humans are complex systems which can be simplified into interconnected compartments. When a drug, especially one based on bacteria, is administered, it doesn't remain static. It moves, interacts, and gets metabolized in various compartments of the body. To encapsulate this dynamic, a multi-compartment model that divides the body into several compartments is proposed, each representing a significant region where the drug might interact or get metabolized.
+Humans are complex systems which can be simplified into interconnected compartments. When a drug, especially one based on bacteria, is administered, it doesn't remain static. It moves, interacts, and gets metabolized in various compartments of the body. To encapsulate this dynamic, a 
+<strong><font color=#8B0012>m</font></strong>ulti-<strong><font color=#8B0012>c</font></strong>ompartment (MC) model that divides the body into several compartments is proposed, each representing a significant region where the drug might interact or get metabolized.
 
 <br>
 
@@ -821,7 +863,7 @@ To simulate the changing trend of compartments when the bacteria shows different
 
 <font size=3 color=grey>
 
-Figure 1-10 | Concentration trend by different elimination rate
+Figure 16-25 | Concentration trend by different elimination rate
 
 </font>
 
@@ -891,7 +933,32 @@ The multi-compartment model offered a profound insight into the journey of a bac
 
 <br>
 
-Harimoto, T., Hahn, J., Chen, YY. *et al.* A programmable encapsulation system improves delivery of therapeutic bacteria in mice. *Nat Biotechnol* **40**, 1259–1269 (2022). https://doi.org/10.1038/s41587-022-01244-y
+1. Harimoto, T., Hahn, J., Chen, YY. *et al.* A programmable encapsulation system improves delivery of therapeutic bacteria in mice. *Nat Biotechnol* **40**, 1259–1269 (2022). https://doi.org/10.1038/s41587-022-01244-y
+
+2. Gao Y, Zhou H, Liu G, Wu J, Yuan Y, Shang A. Tumor Microenvironment: Lactic Acid Promotes Tumor Development. J Immunol Res. 2022 Jun 12;2022:3119375. doi: 10.1155/2022/3119375. PMID: 35733921; PMCID: PMC9207018.      
+
+3.  Brizel D. M., Schroeder T., Scher R. L., et al. Elevated tumor lactate concentrations predict for an increased risk of metastases in head-and-neck cancer. International Journal of Radiation Oncology • Biology • Physics . 2001;51(2):349–353. doi: 10.1016/S0360-3016(01)01630-3.
+
+4. Bintu L, Buchler NE, Garcia HG, Gerland U, Hwa T, Kondev J, Phillips R. Transcriptional regulation by the numbers: models. Curr Opin Genet Dev. 2005 Apr;15(2):116-24. doi: 10.1016/j.gde.2005.02.007. PMID: 15797194; PMCID: PMC3482385.
+
+5. Mettert EL, Kiley PJ. Reassessing the Structure and Function Relationship of the O2 Sensing Transcription Factor FNR. Antioxid Redox Signal. 2018 Dec 20;29(18):1830-1840. doi: 10.1089/ars.2017.7365. Epub 2017 Nov 14. PMID: 28990402; PMCID: PMC6217745
+
+6. Morrison M, Razo-Mejia M, Phillips R. Reconciling kinetic and thermodynamic models of bacterial transcription. PLoS Comput Biol. 2021 Jan 19;17(1):e1008572. doi: 10.1371/journal.pcbi.1008572. PMID: 33465069; PMCID: PMC7845990.
+
+7. Muller-Hill, B., 1996. The Lac Operon, first ed. de Gruyter, Berlin.
+Narang, A., Pilyugin, S.S., 2006. Why does the lac exhibit no bistability
+during growth of Escherichia coli on lactose or lactose + glucose? Bull.
+Math. Biol., submitted for publication.
+
+8. Jensen D, Galburt EA. The Context-Dependent Influence of Promoter Sequence Motifs on Transcription Initiation Kinetics and Regulation. J Bacteriol. 2021 Mar 23;203(8):e00512-20. doi: 10.1128/JB.00512-20. PMID: 33139481; PMCID: PMC8088511.
+
+9. Wang J, Zhang S, Lu H, Xu H. Differential regulation of alternative promoters emerges from unified kinetics of enhancer-promoter interaction. Nat Commun. 2022 May 17;13(1):2714. doi: 10.1038/s41467-022-30315-6. PMID: 35581264; PMCID: PMC9114328.
+
+10. Oehler S, Amouyal M, Kolkhof P, von Wilcken Bergmann B, Hill BM. Quality and position of the three lac operators of E. coli define efficiency of repression. The EMBO journal. 1994; 13(14):3348–3355. https://doi.org/10.1002/j.1460-2075.1994.tb06637.x PMID: 8045263
+
+11. Oehler S, Eismann ER, Kra¨mer H, Hill BM. The three operators of the lac operon cooperate in repression. The EMBO journal. 1990; 9(4):973–979. https://doi.org/10.1002/j.1460-2075.1990.tb08199.x PMID: 2182324
+
+12. Bintu L, Buchler NE, Garcia HG, Gerland U, Hwa T, Kondev J, Kuhlman T, Phillips R. Transcriptional regulation by the numbers: applications. Curr Opin Genet Dev. 2005 Apr;15(2):125-35. doi: 10.1016/j.gde.2005.02.006. PMID: 15797195; PMCID: PMC3462814.
 
 <br>
 
diff --git a/pages/index.vue b/pages/index.vue
index 9f914c7b902c04b592c92b6ae9c26d382e3144d9..28eae2bab14c162f1ecb01285babc953b152fa4c 100644
--- a/pages/index.vue
+++ b/pages/index.vue
@@ -29,7 +29,7 @@
             <span class="block lg:pt-[2vh] lg:pb-[2vh] pt-[1vh] pb-[1vh]" style="">five-year survival rate.</span>
           </div>
         </h5>
-        <img class="iFlash absolute" src="https://static.igem.wiki/teams/4713/wiki//index/cancer-live.png" alt="pancreatic cancer" style="width: 50%; top: 10%; left:48%">
+        <img class="iFlash absolute Rotate" src="https://static.igem.wiki/teams/4713/wiki//index/cancer-live.png" alt="pancreatic cancer" style="width: 50%; top: 10%; left:48%">
       </section>
     </div>
     <div id="Pancreatic Cancer2" class="h-screen z-10" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/background2.png) center center / cover">
@@ -132,74 +132,73 @@
     </div>
 
     <div class=" h-screen" id="Our Project3" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/background2.png) center center / cover">
-      <h2 class="wordfrom text-center mx-auto pt-10 font-bold text-3xl lg:text-5xl" style="">Concerns and Solutions</h2>
-      <div class="wordfrom text-center mx-auto pt-[8vh]" style="display: grid; grid-template-columns: 33% 33% 33%; align-items: center; grid-column-gap: 0px; font-size: 18px; max-width: 2000px;">
-          <div class="card mx-auto">
-            <div class="card-inner">
-              <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover ">
-                <strong><p style="margin: 0;">Effectiveness of Targeting</p></strong>
-              </div>
-              <div class="card-back" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/assassin.png) center center / cover">
-                Effectiveness of Targeting
+            <h2 class="wordfrom text-center mx-auto pt-10 font-bold" style="font-size: 40px;">Concerns and Solutions</h2>
+            <div class="wordfrom text-center mx-auto pt-[8vh]" style="display: grid; grid-template-columns: 33% 33% 33%; align-items: center; grid-column-gap: 0px; font-size: 18px; max-width: 2000px;">
+                <div class="card mx-auto">
+                  <div class="card-inner">
+                    <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover ;font-size: 40px;">
+                      <p style="margin: 0;">Effectiveness <br><br> of <br><br> Targeting</p >
+                    </div>
+                    <div class="card-back" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/assassin.png) center center / cover">
+                      Effectiveness of Targeting
+                      <br>
+                      <br>
+                      ✔ We designed a hybrid promoter functioning as an AND-gate
+                      <br>
+                      <br>
+                      ✔ Concerning both high lactate and hypoxicity
+                      <br>
+                      <br>
+                      ✔ Lower load compared to traditional AND-gate
+                    </div>
+                  </div>
+                </div>
+                <div class="card mx-auto">
+                  <div class="card-inner">
+                    <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover;font-size: 40px;">
+                      <p style="margin: 0;">Intravenous <br><br> Safety</p >
+                    </div>
+                    <div class="card-back" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/assassin.png) center center / cover">
+                      Intravenous safety
+                      <br>
+                      <br>
+      ✔ We introduced controllable enveloping concept and designed an MCM model to imitate.
                 <br>
                 <br>
-                We designed a hybrid promoter functioning as an AND-gate
+                ✔ Wet lab solution
                 <br>
                 <br>
-                Concerning both high lactate and hypoxicity
+                ✔ Dry lab imitation
                 <br>
                 <br>
-                Lower load compared to traditional AND-gate
+                ✔ A safe strain
               </div>
             </div>
           </div>
           <div class="card mx-auto">
             <div class="card-inner">
-              <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover">
-                <strong><p style="margin: 0;">Intravenous safety</p></strong>
-              </div>
-              <div class="card-back" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/assassin.png) center center / cover">
-                Intravenous safety
-                <br>
-                <br>
-                We introduced controllable enveloping concept and designed an MCM model to imitate.
-                <br>
-                <br>
-                Wet lab solution
-                <br>
-                <br>
-                Dry lab imitation
-                <br>
-                <br>
-                A safe strain
-              </div>
-            </div>
-          </div>
-          <div class="card mx-auto">
-            <div class="card-inner">
-              <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover">
-                <p class="font-bold" style="margin: 0">Social Concerns</p>
+              <div class="card-front" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/king3.png) center center / cover;font-size: 40px;">
+                <p style="margin: 0">Social <br><br> Concerns</p >
               </div>
               <div class="card-back" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/assassin.png) center center / cover">
                 Social Concerns
                 <br>
                 <br>
-                We launched abundant human practices and educations.
+                ✔ We launched abundant human practices and educations.
                 <br>
                 <br>
-                Professional opinions
+                ✔ Professional opinions
                 <br>
                 <br>
-                Patients' experiences
+                ✔ Patients' experiences
                 <br>
                 <br>
-                Middle school and College education
+                ✔ Middle school and College education
               </div>
             </div>
         </div>
-      </div>  
     </div>
-
+  </div>
   </div>
   <div class="h-screen" id="over" style="background: url(https://static.igem.wiki/teams/4713/wiki/index/background2.png) center center / cover">
     <img src="https://static.igem.wiki/teams/4713/wiki/index/right-1.png" alt="right" style="width: 20% ;right: 0%" class="top-[4rem] sticky float-right arrow z-10" >
@@ -442,7 +441,8 @@ onMounted(() => {
 
   let Rotate = gsap.utils.toArray('.Rotate')
   Rotate.forEach(a=>{
-    gsap.to(a, {duration: 1.5, rotation: 360, repeat: -1,  ease: "none"});
+    let ro = gsap.timeline()
+      ro.to(a, {duration: 1.5, rotation: 30, repeat: -1, yoyo: true,  ease: "none"})
   })
 
   const typewriter = document.querySelector('.typewriter');