Human activity recognition for analyzing stress behavior based on Bi-LSTM

BACKGROUND

Stress is one of the critical health factors that could be detected by Human Activity Recognition (HAR) which consists of physical and mental health. HAR can raise awareness of self-care and prevent critical situations. Recently, HAR used non-invasive wearable physiological sensors. Moreover, deep learning techniques are becoming a significant tool for analyzing health data.

OBJECTIVE

In this paper, we propose a human lifelog monitoring model for stress behavior recognition based on deep learning, which analyses stress levels during activity. The proposed approach considers activity and physiological data for recognizing physical activity and stress levels.

METHODS

To tackle these issues, we proposed a model that utilizes hand-crafted feature generation techniques compatible with a Bidirectional Long Short-Term Memory (Bi-LSTM) based method for physical activity and stress level recognition. We have used a dataset called WESAD, collected using wearable sensors for model evaluation. This dataset presented four levels of stress emotion, including baseline, amusement, stress, and meditation.

RESULTS

The following results are from the hand-crafted feature approaches compatible with the bidirectional LSTM model. The proposed model achieves an accuracy of 95.6% and an F1-score of 96.6%.

CONCLUSION

The proposed HAR model efficiently recognizes stress levels and contributes to maintaining physical and mental well-being.

Prediction of trehalose-metabolic pathway and comparative analysis of KEGG, MetaCyc, and RAST databases based on complete genome of Variovorax sp. PAMC28711

BACKGROUND

Metabolism including anabolism and catabolism is a prerequisite phenomenon for all living organisms. Anabolism refers to the synthesis of the entire compound needed by a species. Catabolism refers to the breakdown of molecules to obtain energy. Many metabolic pathways are undisclosed and many organism-specific enzymes involved in metabolism are misplaced. When predicting a specific metabolic pathway of a microorganism, the first and foremost steps is to explore available online databases. Among many online databases, KEGG and MetaCyc pathway databases were used to deduce trehalose metabolic network for bacteria Variovorax sp. PAMC28711. Trehalose, a disaccharide, is used by the microorganism as an alternative carbon source.

RESULTS

While using KEGG and MetaCyc databases, we found that the KEGG pathway database had one missing enzyme (maltooligosyl-trehalose synthase, EC 5.4.99.15). The MetaCyc pathway database also had some enzymes. However, when we used RAST to annotate the entire genome of Variovorax sp. PAMC28711, we found that all enzymes that were missing in KEGG and MetaCyc databases were involved in the trehalose metabolic pathway.

CONCLUSIONS

Findings of this study shed light on bioinformatics tools and raise awareness among researchers about the importance of conducting detailed investigation before proceeding with any further work. While such comparison for databases such as KEGG and MetaCyc has been done before, it has never been done with a specific microbial pathway. Such studies are useful for future improvement of bioinformatics tools to reduce limitations.

GCRNN: graph convolutional recurrent neural network for compound–protein interaction prediction

Background

Compound–protein interaction prediction is necessary to investigate health regulatory functions and promotes drug discovery. Machine learning is becoming increasingly important in bioinformatics for applications such as analyzing protein-related data to achieve successful solutions. Modeling the properties and functions of proteins is important but challenging, especially when dealing with predictions of the sequence type.

Result

We propose a method to model compounds and proteins for compound–protein interaction prediction. A graph neural network is used to represent the compounds, and a convolutional layer extended with a bidirectional recurrent neural network framework, Long Short-Term Memory, and Gate Recurrent unit is used for protein sequence vectorization. The convolutional layer captures regulatory protein functions, while the recurrent layer captures long-term dependencies between protein functions, thus improving the accuracy of interaction prediction with compounds. A database of 7000 sets of annotated compound protein interaction, containing 1000 base length proteins is taken into consideration for the implementation. The results indicate that the proposed model performs effectively and can yield satisfactory accuracy regarding compound protein interaction prediction.

Conclusion

The performance of GCRNN is based on the classification accordiong to a binary class of interactions between proteins and compounds The architectural design of GCRNN model comes with the integration of the Bi-Recurrent layer on top of CNN to learn dependencies of motifs on protein sequences and improve the accuracy of the predictions.