Prediction of Hormone-Binding Proteins Based on K-mer Feature Representation and Naive Bayes

Protein function annotation and virulence factor identification of Klebsiella pneumoniae genome by multiple machine learning models

Klebsiella pneumoniae is a type of Gram-negative bacterium which can cause a range of infections in human. In recent years, an increasing number of strains of K. pneumoniae resistant to multiple antibiotics have emerged, posing a significant threat to public health. The protein function of this bacterium is not well known, thus a systematic investigation of K. pneumoniae proteome is in urgent need. In this study, the protein functions of this bacteria were re-annotated, and their function groups were analyzed. Moreover, three machine learning models were built to identify novel virulence factors. Results showed that the functions of 16 uncharacterized proteins were first annotated by sequence alignment. In addition, K. pneumoniae proteins share a high proportion of homology with Haemophilus influenzae and a low homology proportion with Chlamydia pneumoniae. By sequence analysis, 10 proteins were identified as potential drug targets for this bacterium. Our model achieved a high accuracy of 0.901 in the benchmark dataset. By applying our models to K. pneumoniae, we identified 39 virulence factors in this pathogen. Our findings could provide novel clues for the treatment of K. pneumoniae infection.

Accurate Prediction of Antifreeze Protein from Sequences through Natural Language Text Processing and Interpretable Machine Learning Approaches

Antifreeze proteins (AFPs) bind to growing iceplanes owing to their structural complementarity nature, thereby inhibiting the ice-crystal growth by thermal hysteresis. Classification of AFPs from sequence is a difficult task due to their low sequence similarity, and therefore, the usual sequence similarity algorithms, like Blast and PSI-Blast, are not efficient. Here, a method combining n-gram feature vectors and machine learning models to accelerate the identification of potential AFPs from sequences is proposed. All these n-gram features are extracted from the K-mer counting method. The comparative analysis reveals that, among different machine learning models, Xgboost outperforms others in predicting AFPs from sequence when penta-mers are used as a feature vector. When tested on an independent dataset, our method performed better compared to other existing ones with sensitivity of 97.50%, recall of 98.30%, and f1 score of 99.10%. Further, we used the SHAP method, which provides important insight into the functional activity of AFPs.

Empirical comparison and recent advances of computational prediction of hormone binding proteins using machine learning methods

Hormone binding proteins (HBPs) belong to the group of soluble carrier proteins. These proteins selectively and non-covalently interact with hormones and promote growth hormone signaling in human and other animals. The HBPs are useful in many medical and commercial fields. Thus, the identification of HBPs is very important because it can help to discover more details about hormone binding proteins. Meanwhile, the experimental methods are time-consuming and expensive for hormone binding proteins recognition. Computational prediction methods have played significant roles in the correct recognition of hormone binding proteins with the use of sequence information and ML algorithms. In this review, we compared and assessed the implementation of ML-based tools in recognition of HBPs in a unique way. We hope that this study will give enough awareness and knowledge for research on HBPs.

Machine learning-aided scoring of synthesis difficulties for designer chromosomes

Designer chromosomes are artificially synthesized chromosomes. Nowadays, these chromosomes have numerous applications ranging from medical research to the development of biofuels. However, some chromosome fragments can interfere with the chemical synthesis of designer chromosomes and eventually limit the widespread use of this technology. To address this issue, this study aimed to develop an interpretable machine learning framework to predict and quantify the synthesis difficulties of designer chromosomes in advance. Through the use of this framework, six key sequence features leading to synthesis difficulties were identified, and an eXtreme Gradient Boosting model was established to integrate these features. The predictive model achieved high-quality performance with an AUC of 0.895 in cross-validation and an AUC of 0.885 on an independent test set. Based on these results, the synthesis difficulty index (S-index) was proposed as a means of scoring and interpreting synthesis difficulties of chromosomes from prokaryotes to eukaryotes. The findings of this study emphasize the significant variability in synthesis difficulties between chromosomes and demonstrate the potential of the proposed model to predict and mitigate these difficulties through the optimization of the synthesis process and genome rewriting.