Search

Diabetes mellitus is a chronic disease that can occur in anyone. Up until now, there are no specific drugs have been found which can completely cure diabetes. One of the possible steps to treat diabetes mellitus is by inhibiting the growth of the Diacylglycerol Acyltransferase-1 (DGAT-1) enzyme. This study aims to build a QSAR model on DGAT-1 inhibitors as anti-diabetic using Particle Swarm Optimization (PSO) and Support Vector Machine (SVM). Acyl-CoA: DGAT1 is a microsomal enzyme in lipogenesis which is increased in metabolically active cells to meet nutrient requirements. Microsomal enzymes that have an important in the triglyceride-synthesis process of 1,2-diacylglycerol by-catalyzing-acyl-coa-dependent-acylations as anti-diabetics. The dataset used in this study consists of 228 samples containing molecular structures and their inhibitor activities. We reduce the number of features by removing features with a standard deviation less than the threshold value, followed by the PSO algorithm. The best-predicted result is obtained through the implementation of SVM with RBF kernel, with the score of and are 0.75 and 0.67, respectively., Diabetes mellitus is a chronic disease that can occurred to anyone. Up until now, there are no specific drugs that have been found which can completely cure diabetes. One of the possible steps to treat diabetes mellitus is by inhibiting the growth of Diacylglycerol Acyltransferase-1 (DGAT-1) enzyme. This study aims to build a QSAR model on DGAT-1 inhibitors as anti-diabetic using Particle Swarm Optimization (PSO) and Support Vector Machine (SVM). Acyl-CoA: DGAT1 is a mikrosomal enzyme in lipogenesis which is increased in metabolically active cells to meet nutrient requirements. Microsomal-enzymes-that have an-important in the triglyceride-synthesis process-of 1,2-diacylglycerol by-catalyzing-acyl-coa-dependent-acylations as anti-diabetics. The dataset used in this study consists of 228 samples containing molecular structures and their inhibitor activities. We reduce the number of features by removing features with a standard deviation less than the threshold value, followed by the PSO algorithm. The best-predicted result is obtained through the implementation of SVM with RBF kernel, with the score of and are 0.75 and 0.67, respectively. &nbsp

Diabetes mellitus is a chronic illness that can affect anyone, while the medicine that can entirely cure diabetes has not been discovered yet. Dipeptidyl Peptidase IV (DPP IV) inhibitor is one of the agents with potency as an anti-diabetic treatment. In this work, we utilized the machine learning method to predict the activity of DPP IV as an anti-diabetic agent. We combined Particle Swarm Optimization (PSO) method for features selection and the Support Vector Machine (SVM) for the prediction model. Three SVM kernels, i.e., radial basis function (RBF), polynomial, and linear, were utilized, and their performance was compared. A Hyperparameter tuning procedure was conducted to improve the performance of models. According to the results, we found that the best model obtained from SVM with RBF kernel with the value R2 of train and test set are 0.79 and 0.85, respectively. &nbsp, Diabetes mellitus is a chronic illness that can affect anyone, while the medicine that can entirely cure diabetes has not been discovered yet. Dipeptidyl Peptidase IV (DPP IV) inhibitor is one of the agents with potency as an anti-diabetic treatment. In this work, we utilized the machine learning method to predict the activity of DPP IV as an anti-diabetic agent. We combined Particle Swarm Optimization (PSO) method for features selection and the Support Vector Machine (SVM) for the prediction model. Three SVM kernels, i.e., radial basis function (RBF), polynomial, and linear, were utilized, and their performance was compared. A Hyperparameter tuning procedure was conducted to improve the performance of models. According to the results, we found that the best model obtained from SVM with RBF kernel with the value R2 of train and test set are 0.79 and 0.85, respectively.

SARS-CoV-2 PLpro was investigated as a therapeutic target for potent antiviral drugs due to its essential role in not only viral replication but also in regulating the inborn immune response. Several computational approaches, including homology modelling, molecular docking, and molecular dynamics (MD) studies, were employed to search for promising drugs in treating SARS-CoV-2. Eighty-one compounds, sub-structurally similar to the antiviral drug, were used as potential inhibitors of PLpro. From our results, three complexes containing the ligands with Pubchem IDs: 153012995, 12149203, and 123608715 showed lower binding energies than the control (Ritonavir), indicating that they may become promising inhibitors for PLpro. MD was performed in a water solvent to validate the stability of the three complexes. All complexes achieved stable structure during the simulation as no significant fluctuations were observed in the validation parameters. Moreover, the binding energy for each complex was estimated using the MM-GBSA method. Complex 1 was the most stable structure based on the lowest binding energy score and its structure remained in a similar cavity with the docket snapshot. Based on our studies, three ligands were assumed to be potential inhibitors. The ligand of complex 1 may become the most promising antiviral drug against SARS-CoV-2 targeting PLpro.

We investigate thiazolyl-pyrazoline derivatives as promising drugs for the anti-malarial. Protein kinase G is a primary target for treating malaria due to its essential role in Plasmodium falciparu...

This research applied a machine learning technique for predicting the water quality parameters of Kelantan River using the historical data collected from various stations. Support Vector Machine (SVM) was used to develop the prediction model. Six water quality parameters (dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD), ammonia nitrogen (NH3-N), and suspended solids (SS)) were predicted. The dataset was obtained from the measurement of 14 stations of Kelantan River from September 2005 to December 2017 with a total sample of 148 monthly data. We defined 3 schemes of prediction to investigate the contribution of the attribute number and the model performance. The outcome of the study demonstrated that the prediction of the suspended solid parameter gave the best performance, which was indicated by the highest values of the R2 score. Meanwhile, the prediction of the COD parameter gave the lowest score of R2 score, indicating the difficulty of the dataset to be modelled by SVM. The analysis of the contribution of attribute number shows that the prediction of the four parameters (DO, BOD, NH3-N, and SS) is directly proportional to the performance of the model. Similarly, the best prediction of the pH parameter is obtained from the utilization of the least number of attributes found in scheme 1. Keywords: machine learning, water quality parameters, turbidity, suspended solids, Kelantan River.

COVID-19 is a type of pneumonia caused by the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2). This virus causes severe acute respiratory syndrome and 2 million active cases of COVID-19 have been found worldwide. A new strain of the SARS-CoV-2 virus emerged that proved to be more virulent than its predecessor. Regarding the design of a new inhibitor for this strain, SARS-CoV Main Protease (Mpro) was used as the target inhibitor. In the in silico development, the Quantitative Structure-Activity Relationship (QSAR) method is commonly used to predict the biological activity of unknown compounds to improve the process of drug design of a disease, including COVID-19. In this study, we aim to develop a QSAR model to predict the activity of aromatic disulfide compounds as SARS-CoV Mpro inhibitors using Genetic Algorithm (GA) – Support Vector Machine (SVM). GA was used for feature selection, while SVM was used for model prediction. The used dataset is set of features of aromatic disulfide compounds, along with information on the toxicity activity. We found that the best SVM model was obtained through the implementation of the polynomial kernel with the value of R2­­train and R2test­ scores are 0.952 and 0.676, respectively.

Social media is one of the ways to connect every individual in the world. It also used by irresponsible people to spread a hoax. Hoax is false news that is made as if it is true. It may cause anxiety and panic in society. It can affect the social and political conditions. This era, the most popular social media is Twitter. It is a place for sharing information and users around the world can share and receive news in short messages or called tweet. Hoax detection gained significant interest in the last decade. Existing hoax detection methods are based on either news-content or social-context using user-based features. In this study, we present a hoax detection based on FF & BP neural networks. In the developing of it, we used two vectorization methods, TF-IDF and Word2Vec. Our model is designed to automatically learn features for hoax news classification through several hidden layers built into the neural network. The neural network is actually using the ability of the human brain that is able to provide stimulation, process, and output. It works by the neuron to process every information that enters, then is processed through a network connection, and will continue learning to produce abilities to do classification. Our proposed model would be helpful to provide a better solution for hoax detection. Data collection obtained through crawling used Twitter API and retrieve data according to the keywords and hashtags. The neural networks highest accuracy obtained using TF-IDF by 78.76%. We also found that data quality affects the performance., Informasi dihasilkan di internet melalui berbagai media sosial. Tidak aneh jika berita hoax sering beredar di media sosial yang telah mempengaruhi kelompok tertentu untuk menjatuhkan seseorang. Berita hoax adalah berita yang masih diragukan. Di era ini, media sosial yang paling populer adalah Twitter. Twitter adalah tempat untuk berbagi informasi, di mana pengguna di seluruh dunia dapat berbagi dan menerima informasi dalam pesan singkat. Dalam penelitian ini, penulis membangun sistem untuk mendeteksi hoax di Twitter. Metode klasifikasi menggunakan feed-forward dan back-propagation neural networks. Validasi data didasarkan pada isi pesan. Setiap pesan melewati pre-processing dan pembobotan kata sebelum diproses oleh sistem klasifikasi. Pembobotan kata menggunakan metode TF-IDF dan Word2Vec sebagai perbandingan hasil kinerja sistem yang dihitung oleh confusion matrix. Dalam mengembangkan sistem, penulis mengumpulkan data dari Twitter dalam bentuk tweet sesuai dengan kata kunci dan tagar yang sedang banyak dibahas di Indonesia. Tujuan dari penelitian ini adalah untuk meminimalkan penyebaran berita hoax di Twitter. Dari hasil pengujian, feed-forward dan back-propagation neural networks dapat digunakan dalam deteksi hoax. Hasil kinerja sistem menunjukkan bahwa akurasi tertinggi dalam pembobotan Word2Vec dengan tokenizer unigram dengan akurasi 77,57%. Word2Vec bekerja lebih cepat dan lebih dalam dibandingkan dengan TF-IDF. Unigram menunjukkan kinerja lebih baik daripada bigram dan trigram karena kombinasi kata-kata menyebabkan banyak noise dalam kata sehingga sistem tidak dapat bekerja secara optimal.

Spirulina platensis contain variety of pigments, such as chlorophylls, carotenoids, and phycocyanin. Phycocyanin (PC) exists in abundance, but due to its instability, the utilization of this pigment is still very limited. In this study, the PC was modified using formaldehyde crosslinks yielding phycocyanin-formaldehyde (PC-F), and its photostability was evaluated. The PC-F formation was designated by the distinctive alterations of the maximum absorption to 611 nm, which was 10 nm blue-shifted than those of the PC. Additionally, the sharp peaks of FTIR spectra at 1636 nm for C O and at 1019 nm for C O C, suggesting the interaction of phycocyanin with formaldehyde. The PC-F showed stabilization improvement up to 1.53-folds after 300 mins of yellow light exposure than those of PC. Contrary to yellow light irradiation, a severe decrease of PC-F absorbance was observed reach to 4.9-folds under UV-B irradiation. The poor stability of PC-F upon white light and UV-A irradiation were indicated by the decline of PC-F absorbance up to 1.72 and 1.80, respectively. Moreover, the present study suggests that the modification of phycocyanin by formaldehyde crosslink can increase photostability upon yellow light irradiation.

Dengue fever is a disease transmitted by infected mosquitoes. This disease spreads in several countries, especially those with a tropical climate. To date, there is no specific drug that can be used to treat dengue. Use of clinically investigated drugs, such as Balapiravir, is still not effective in inhibiting the activity of virus replication. The design of a drug candidate can be performed by using the non-structural protein 3 (NS3) as target. This study aimed to develop QSAR models to predict the inhibitory activity class of NS3 inhibitors. The classification was performed by using feature importance analysis for selecting the descriptors and three ensemble methods, i.e. random forest (RF), adaptive boosting (AdaBoost), and extremely randomized trees (ERT), for model design and prediction. Hyperparameter tuning was performed to improve the performance of the models. Based on the results, we found that model 9, developed from ERT produced the best performance with values of accuracy and AUC equal to 0.73 and 0.82, respectively. Use of y-scrambling method allowed us to confirm that the model was not related to the chance correlation.

Cancer is a disease that occurs due to the uncontrolled growth of abnormal cells causing body tissue damage. This disease is considered as a deadly disease. In 2019, 1700 deaths occur every day due to cancer. Some effective anti-cancer agents are known to cause temporary to chronic toxic effects. There are several compounds that have the potential to become anticancer drugs, one of them is indenopyrazole. Recently, not many QSAR studies have been conducted to exploit the potential of indenopyrazole as anti-cancer agent. The goal of this research is to implement simulated annealing and support vector machine method in the QSAR study to predict the activity of indenopyrazole derivatives as anti-cancer drugs. Simulated annealing is used for feature selection and support vector machine is used for model development. In this research, we used three kernel models for SVM, namely SVM with RBF kernel, SVM with linear kernel, and SVM with the polynomial kernel. From three models that were regressed, SVM with RBF kernel has parameter C=10, gamma=scale and epsilon=0.1 produce R2 score train and test 0.79 and 0.60, respectively. SVM with linear kernel has parameter C=1000, degree=1 and epsilon=0.1 produce R2 score train and test 0.61 and 0.63, respectively. SVM with polynomial kernel has parameter C=1000, degree=2 and epsilon=0.1 produce R2 score train and test 0.72 and 0.50, respectively. Based on the validation results, only model with RBF kernel which parameters satisfy all the criteria. From the result we can conclude that the model with RBF kernel is the best model and acceptable.

Cyanobacteria such as Spirulina platensis secretes numerous biomolecules while consuming CO

Weather prediction is usually performed for a reference in planning future activity. The prediction is performed by considering several parameters, such as temperature, air pressure, humidity, wind, rainfall, and others. In this study, the temperature, as one of weather parameters, is predicted by using time series from January 2015 to December 2017. The data was obtained from Lembaga Ilmu Pengetahuan Indonesia (LIPI) weather measurement station in Muaro Anai, Padang. The predictions were carried out by using Convolutional Neural Network (CNN), Multilayer Perceptron (MLP), and the hybrid of CNN-MLP methods. The parameters used in the CNN method, such as the number of filters and kernel size, and used in the MLP method, such as the number of hidden layers and number of neurons, were selected by performing the hyperparameter tuning procedure. After obtaining the best parameters for both methods, the performance of both methods was evaluated by calculating the value of Root Mean Square Error (RMSE) and R2. Based on the results, we found that the prediction by CNN is more accurate than other method. This is indicated by the highest value of R2 of the prediction obtained by CNN method.

Malaria is a disease caused by the Plasmodium falciparum parasite and leads to many cases of deaths. Recently, the combination of several drugs has been used to treat this disease. However, the parasite is known to be resistant to the anti-malarial agent. Hence, a new candidate for an anti-malarial drug is required to solve the resistance problem. One compound that is promising as an anti-malarial agent is fusidic acid derivatives. Fusidic acid is an antibiotic that is work by preventing parasite growth. Besides, fusidic acid is known to have antiplasmodial activity although the IC50 is still poor. However, the activity can be improved by optimizing the structure through its derivatives. In this study, we developed a QSAR model to predict the activity of fusidic acid derivatives as anti-malarial agent. The model was developed by using Simulated Annealing (SA) for feature selection and Support Vector Machine (SVM) for model development. The results show that SA produces a satisfying combination of features that are indicated by the trend of MSE value during the selection process. Regarding the performance, SVM with RBF kernel produces the best result of the validation parameter. This indicates that the model is valid to be used to predict a compound with unknown activity values for anti-malarial agents.

In the last decade, Bandung has become one of the tourism destination places in Indonesia. It is reported that almost 6.7 million visitors come to Bandung in 2018, and the number increased by almost 4% per year since 2014. This rise in the number of visitors leads to the establishment of several toll gates as main access points throughout the city. As one of the busiest ones, Buah Batu toll gate is frequently congested because of the location that is close to the southern part of Bandung. To overcome this problem, a traffic regulation based on computer simulation is urgently required. In this study, we simulate the traffic system on the Buah Batu toll gate by using a combination of Nagel-Schreckenberg (NaSch) and Daoudia and Moussa (DM) models. NaSch model was used to defined vehicle movement, while the DM model was used to allow a vehicle to change lane. We defined three scenarios to evaluate the effectivity of the closing gate scheme. We found that the closing of gate 5 is more effective than the closing of gate 1. We also investigated the contribution of traffic density and driver’s behavior, e.g., stopping behavior and lane-changing behavior, to the average velocity of the vehicles.

Employees' qualities affect companies' performances and with a large number of applicants, it's difficult to find suitable applicants. To help with it, companies carry out psychological tests to know applicants' personalities, since personality's considered to have a relationship with work performances. But psychological testing requires a lot of effort, cost, and human resources. Thus with a system that can classify personalities through text can help reduce the effort needed. Similar studies carried out with the big five personalities as the theoretical basis and used one of the personality traits, namely using the k-NN method with 65% accuracy. Based on these studies, accuracy can improve by finding the best parameters using all of the big five personalities. This research is conducted based on the big five personality traits and related traits, namely consciousness and agreeableness. The data used is text data that's been labelled, pre-processed and feature selected. The clean text data is used to create a classification model using multinomial Naive Bayes and decision trees. There are 6 models built based on 3 work cultures, decision tree with an accuracy of 33%, 66%, 80%, and multinomial naïve Bayes with an accuracy of 83%, 50%, 60%, which resulted as better performance.

Selecting the best talents from a large number of job applicants is challenging, especially for big companies that usually receive tens of thousands of applicants for every job opening. One of the most costly and time-consuming applicant selection stages is the interview process, since it usually performs face to face meetings and involves third parties to do the interviews and analyze the result. To this end, Human Capital Directorate at Telkom Indonesia adopts AI technology to automate some stages of job applicant selection to reduce manual process and third-party involvement. In this paper, we investigate appropriate feature extraction methods to automate job interview grading for reducing bias and human errors. TFIDF, one of the most popular feature extractions, is compared with word embedding to find the optimal method and parameters in classifying interview verbatims with ANN classifier. Based on the test results, the average accuracy for TFIDF outperforms word embedding by 85.22% against 74.88%, respectively. Therefore, for the case of job interview grading using our dataset, TF-IDF performs better to reduce the number of dimensions.

Malaria is a disease that caused many adverse effects on humans. Various attempts have been done to find new anti-malarial agents due to the resistance problem of the existing drug. Fusidic acid is known as one of a compound that is promising to be used as an anti-malaria agent. However, this compound should be derived to obtain a new fusidic acid derivative that has better activity. The exploration of the compound in conventional style has a shortcoming in the term of time and cost. Therefore, an alternative method is required to accelerate the design. In this study, we applied a quantitative structure-activity relationship (QSAR) to produce a predictive model. The produced model can be used to predict the activity of the compound as an anti-malaria agent. The development of the model was performed by using genetic algorithm (GA) for feature selection and artificial neural network (ANN) for model development. We developed five models by utilizing a different number of the descriptor in each model. The validation process was performed by evaluating several validation parameters, such as accuracy. According to the results, we found that the model 3, which is comprised of seven descriptors, produce a better result with the accuracies of internal and external data set are 0.96 and 0.92, respectively.

We investigated the effect of an axial ligand on the electronic properties of type I copper (T1Cu) protein. Several density functional theory (DFT) functionals were utilized, including pure general...

We present a simple coarse-grained model with the molecular crowding effect in solvent to investigate the structure and dynamics of protein complexes including association and/or dissociation processes and investigate some physical properties such as the structure and the reaction rate from the viewpoint of the hydrophobic intermolecular interactions of protein complex. In the present coarse-grained model, a function depending upon the density of hydrophobic amino acid residues in a binding area of the complex is introduced, and the function involves the molecular crowding effect for the intermolecular interactions of hydrophobic amino acid residues between proteins. We propose a hydrophobic intermolecular potential energy between proteins by using the density-dependent function. The present coarse-grained model is applied to the complex of cytochrome f and plastocyanin by using the Langevin dynamics simulation to investigate some physical properties such as the complex structure, the electron tra...