NCTU_Formosa 2017 had completed a Peptide prediction model which can predict peptides for new function. In the model, they used Scoring Card Method (SCM) for machine learning. This year, NCTU_Formosa 2018 continued to use the same method for predicting antimicrobial peptide, in order to seek more candidates for our project.
Difference between revisions of "Team:NCTU Formosa/Dry Lab/Peptide Prediction"
Terrychan999 (Talk | contribs) |
|||
| (5 intermediate revisions by 2 users not shown) | |||
| Line 28: | Line 28: | ||
overflow: auto; | overflow: auto; | ||
font-family: Levenim MT; | font-family: Levenim MT; | ||
| + | padding-bottom: 150px; | ||
} | } | ||
| Line 193: | Line 194: | ||
.card{ | .card{ | ||
| − | width: | + | width: 70%; |
| − | margin-left: | + | margin-left: 17%; |
} | } | ||
| Line 210: | Line 211: | ||
width: 60%; | width: 60%; | ||
margin-left: 20%; | margin-left: 20%; | ||
| + | } | ||
| + | |||
| + | .GitHub:hover{ | ||
| + | transform: scale(1.2); | ||
| + | transition-duration: 0.5s; | ||
} | } | ||
| Line 285: | Line 291: | ||
</p> | </p> | ||
</div> | </div> | ||
| − | <img src="https://static.igem.org/mediawiki/2018/ | + | <img src="https://static.igem.org/mediawiki/2018/5/55/T--NCTU_Formosa--Scoring_card.png" class="card"> |
<div class="explanation"> | <div class="explanation"> | ||
<svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> | <svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> | ||
| − | Figure 1: Scoring card of antimicrobial peptide. The more blue the dipeptide the less likely it is antimicrobial, the more red the dipeptide, the more likely it is antimicrobial. | + | Figure 1: Scoring card of antimicrobial peptide.<br> |
| − | + | <p>The more blue the dipeptide the less likely it is antimicrobial, the more red the dipeptide, the more likely it is antimicrobial.</p> | |
</div> | </div> | ||
</div> | </div> | ||
| Line 295: | Line 301: | ||
<div class="explanation"> | <div class="explanation"> | ||
<svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> | <svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> | ||
| − | Figure 2: The distribution between positive and negative datasets. The smaller the | + | Figure 2: The distribution between positive and negative datasets.<br> |
| − | <div | + | <p>The smaller the overlap area, the more precise the peptide prediction model is.</p> |
| + | </div> | ||
<img src="https://static.igem.org/mediawiki/2018/b/b4/T--NCTU_Formosa--ROC_curve.png" class="ROC"> | <img src="https://static.igem.org/mediawiki/2018/b/b4/T--NCTU_Formosa--ROC_curve.png" class="ROC"> | ||
| − | <svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> | + | <div class="explanation"> |
| − | Figure 3: The | + | <svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-up" class="svg-inline--fa fa-arrow-circle-up fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M8 256C8 119 119 8 256 8s248 111 248 248-111 248-248 248S8 393 8 256zm143.6 28.9l72.4-75.5V392c0 13.3 10.7 24 24 24h16c13.3 0 24-10.7 24-24V209.4l72.4 75.5c9.3 9.7 24.8 9.9 34.3.4l10.9-11c9.4-9.4 9.4-24.6 0-33.9L273 107.7c-9.4-9.4-24.6-9.4-33.9 0L106.3 240.4c-9.4 9.4-9.4 24.6 0 33.9l10.9 11c9.6 9.5 25.1 9.3 34.4-.4z"></path></svg> |
| + | Figure 3: The receiver operating characteristic curve of the peptide prediction model. | ||
| + | </div> | ||
| + | <div class="title_1"><p>Result</p></div> | ||
| + | <div class="text"> | ||
| + | <p> | ||
| + | Finally, our peptide pridiction model help us to filter out six potential antimicrobial peptide in the table below. In the future, with this model, we can promote our system to further usage, such as pathogentic disease and so on. | ||
| + | </p> | ||
</div> | </div> | ||
<div class="table"> | <div class="table"> | ||
| Line 306: | Line 320: | ||
<p class="explanation"> | <p class="explanation"> | ||
<svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-down" class="svg-inline--fa fa-arrow-circle-down fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-143.6-28.9L288 302.6V120c0-13.3-10.7-24-24-24h-16c-13.3 0-24 10.7-24 24v182.6l-72.4-75.5c-9.3-9.7-24.8-9.9-34.3-.4l-10.9 11c-9.4 9.4-9.4 24.6 0 33.9L239 404.3c9.4 9.4 24.6 9.4 33.9 0l132.7-132.7c9.4-9.4 9.4-24.6 0-33.9l-10.9-11c-9.5-9.5-25-9.3-34.3.4z"></path></svg> | <svg class="icon" aria-hidden="true" data-prefix="fas" data-icon="arrow-circle-down" class="svg-inline--fa fa-arrow-circle-down fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M504 256c0 137-111 248-248 248S8 393 8 256 119 8 256 8s248 111 248 248zm-143.6-28.9L288 302.6V120c0-13.3-10.7-24-24-24h-16c-13.3 0-24 10.7-24 24v182.6l-72.4-75.5c-9.3-9.7-24.8-9.9-34.3-.4l-10.9 11c-9.4 9.4-9.4 24.6 0 33.9L239 404.3c9.4 9.4 24.6 9.4 33.9 0l132.7-132.7c9.4-9.4 9.4-24.6 0-33.9l-10.9-11c-9.5-9.5-25-9.3-34.3.4z"></path></svg> | ||
| − | Table 1: The prediction result | + | Table 1: The prediction result from scoring card. |
</p> | </p> | ||
</caption> | </caption> | ||
| Line 347: | Line 361: | ||
</table> | </table> | ||
</div> | </div> | ||
| + | <div class="text"> | ||
| + | <p> | ||
| + | If you are interesting in our model, please click the icon to find out how them work in GitHub! | ||
| + | </p> | ||
| + | </div> | ||
| + | <a href="https://github.com/terrychan999/NCTU_Formosa-igem2018"><img src="https://static.igem.org/mediawiki/2018/1/11/T--NCTU_Formosa--Git_Hub.png" class="GitHub" style="width: 5%; margin-left: 47.5%;"></a> | ||
</div> | </div> | ||
Latest revision as of 00:44, 8 December 2018
Prediction Process
We started from the Uniprot to pick out the peptides with the key word of antimicrobial. After deleting peptides with similarity over 50% with each other, there were totally 425 sequences left. These data will be taken as positive data for our model.
For negative data, we chose the sequences which are not antimicrobial peptides and the sequences length should be around 30 to 300 amino acids. The amount of these random selection is the same as the amount of positive datasets. Therefore, the ratio between positive and negative would be 1:1.
Next, all the datasets were mixed together, 2/3 for training and 1/3 for testing. After training, we got a threshold and the score of each amino acid sequence. Then we can calculate the score of any peptide through this scoring card.
At last, comparing the score of peptides with threshold, we can easily determine whether the unknown peptide might have the function of inhibiting bacteria’s growth.
Figure 1: Scoring card of antimicrobial peptide.
The more blue the dipeptide the less likely it is antimicrobial, the more red the dipeptide, the more likely it is antimicrobial.
Figure 2: The distribution between positive and negative datasets.
The smaller the overlap area, the more precise the peptide prediction model is.
Figure 3: The receiver operating characteristic curve of the peptide prediction model.
Result
Finally, our peptide pridiction model help us to filter out six potential antimicrobial peptide in the table below. In the future, with this model, we can promote our system to further usage, such as pathogentic disease and so on.
Bacteriocin |
Score |
|---|---|
Leucocyclicin Q |
438.38 |
Enterocin B |
464.36 |
Enterocin 96 |
464.06 |
Lacticin Z |
450.92 |
Bovicin HJ50 |
459.87 |
Durancin TW-49M |
478.97 |
Threshold |
431.65 |
If you are interesting in our model, please click the icon to find out how them work in GitHub!
