Cancer-beating molecules (CBMs) are abundant in many types of food and potentially anti-cancer therapeutic agents. In the previous work, researchers introduced a network-based machine learning platform to identify the...Cancer-beating molecules (CBMs) are abundant in many types of food and potentially anti-cancer therapeutic agents. In the previous work, researchers introduced a network-based machine learning platform to identify the cancer-beating molecules, for example,?comparing the similarities in the molecular network between approved anticancer drug and food molecules. Herein, we aim to build on this work to enhance the accuracy of predicting food molecules. In this project, we improve supervised learning approaches by applying Soft Voting algorithm to seven machine learning algorithms: Support Vector Machine with Radial Basis Function (SVM with RBF kernel), multilayer perceptron neural network?(MLP), Random forest, Decision trees,?Gaussian Naive Bayes, Adaboosting, and Bagging. As a result, the accuracy in the dataset of 50 food molecules utilized increased from 82% to 87%, achieving a significant improvement in the precision of?predicting anti-cancer molecules.展开更多
The DNA sequences of an organism play an important influence on its transcription and translation process, thus affecting its protein production and growth rate. Due to the com-plexity of DNA, it was extremely difficu...The DNA sequences of an organism play an important influence on its transcription and translation process, thus affecting its protein production and growth rate. Due to the com-plexity of DNA, it was extremely difficult to predict the macroscopic characteristics of or-ganisms. However, with the rapid development of machine learning in recent years, it be-comes possible to use powerful machine learning algorithms to process and analyze biolog-ical data. Based on the synthetic DNA sequences of a specific microbe, <em>E. coli</em>, I designed a process to predict its protein production and growth rate. By observing the properties of a data set constructed by previous work, I chose to use supervised learning regressors with encoded DNA sequences as input features to perform the predictions. After comparing different encoders and algorithms, I selected three encoders to encode the DNA sequences as inputs and trained seven different regressors to predict the outputs. The hy-per-parameters are optimized for three regressors which have the best potential prediction performance. Finally, I successfully predicted the protein production and growth rates, with the best <em>R</em><sup><em>2</em></sup> score 0.55 and 0.77, respectively, by using encoders to catch the potential fea-tures from the DNA sequences.展开更多
文摘Cancer-beating molecules (CBMs) are abundant in many types of food and potentially anti-cancer therapeutic agents. In the previous work, researchers introduced a network-based machine learning platform to identify the cancer-beating molecules, for example,?comparing the similarities in the molecular network between approved anticancer drug and food molecules. Herein, we aim to build on this work to enhance the accuracy of predicting food molecules. In this project, we improve supervised learning approaches by applying Soft Voting algorithm to seven machine learning algorithms: Support Vector Machine with Radial Basis Function (SVM with RBF kernel), multilayer perceptron neural network?(MLP), Random forest, Decision trees,?Gaussian Naive Bayes, Adaboosting, and Bagging. As a result, the accuracy in the dataset of 50 food molecules utilized increased from 82% to 87%, achieving a significant improvement in the precision of?predicting anti-cancer molecules.
文摘The DNA sequences of an organism play an important influence on its transcription and translation process, thus affecting its protein production and growth rate. Due to the com-plexity of DNA, it was extremely difficult to predict the macroscopic characteristics of or-ganisms. However, with the rapid development of machine learning in recent years, it be-comes possible to use powerful machine learning algorithms to process and analyze biolog-ical data. Based on the synthetic DNA sequences of a specific microbe, <em>E. coli</em>, I designed a process to predict its protein production and growth rate. By observing the properties of a data set constructed by previous work, I chose to use supervised learning regressors with encoded DNA sequences as input features to perform the predictions. After comparing different encoders and algorithms, I selected three encoders to encode the DNA sequences as inputs and trained seven different regressors to predict the outputs. The hy-per-parameters are optimized for three regressors which have the best potential prediction performance. Finally, I successfully predicted the protein production and growth rates, with the best <em>R</em><sup><em>2</em></sup> score 0.55 and 0.77, respectively, by using encoders to catch the potential fea-tures from the DNA sequences.
