Vol 19, No 2 (2024)

Life Sciences

Investigation of LncRNAs Expression as a Potential Biomarker in the Diagnosis and Treatment of Human Brucellosis

Khaledi M., Haddadi M., Aziziraftar S., Neamati F., Sahebkar A., Kodori M., Abavisani M., Fathizadeh H.

Abstract

Long non-coding RNAs (LncRNAs) are significant contributors to bacterial infections and host defense responses, presenting a novel class of gene regulators beyond conventional protein-coding genes. This narrative review aimed to explore the involvement of LncRNAs as a potential biomarker in the diagnosis and treatment of bacterial infections, with a specific focus on Brucella infections. A comprehensive literature review was conducted to identify relevant studies examining the roles of LncRNAs in immune responses during bacterial infections, with a specific emphasis on Brucella infections. Pub- Med, Scopus and other major scientific databases were searched using relevant keywords. LncRNAs crucially regulate immune responses to bacterial infections, influencing transcription factors, proinflammatory cytokines, and immune cell behavior, with both positive and negative effects. The NF-κB pathway is a key regulator for many LncRNAs in bacterial infections. During Brucella infections, essential LncRNAs activate the innate immune response, increasing proinflammatory cytokine production and immune cell differentiation. LncRNAs are associated with human brucellosis, holding promise for screening, diagnostics, or therapeutics. Further research is needed to fully understand LncRNAs' precise functions in Brucella infection and pathogenesis. Specific LncRNAs, like IFNG-AS1 and NLRP3, are upregulated during brucellosis, while others, such as Gm28309, are downregulated, influencing immunosuppression and bacterial survival. Investigating the prognostic and therapeutic potential of Brucellarelated LncRNAs warrants ongoing investigation, including their roles in other immune cells like macrophages, dendritic cells, and neutrophils responsible for bacterial clearance. Unraveling the intricate relationship between LncRNAs and brucellosis may reveal novel regulatory mechanisms and LncRNAs' roles in infection regulation, expediting diagnostics and enhancing therapeutic strategies against Brucella infections.

Current Bioinformatics. 2024;19(2):103-118
pages 103-118 views

TumorDet: A Breast Tumor Detection Model Based on Transfer Learning and ShuffleNet

Zhang T., Pan L., Yang Q., Yang G., Han N., Qiao S.

Abstract

Background:Breast tumor is among the most malignant tumors and early detection can improve patient’s survival rate. Currently, mammography is the most reliable method for diagnosing breast tumor because of high image resolution. Because of the rapid development of medical and artificial intelligence techniques, computer-aided diagnosis technology can greatly improve the detection accuracy of breast tumors and medical imaging has begun to use deep-learning-based approaches. In this study, the TumorDet model is proposed to detect the benign and malignant lesions of breast tumor, which has positive significance for assisting doctors in diagnosis.

Objective:We use the proposed TumorDet to analyze and predict breast tumors on the real MRI dataset.

Methods:(1) We introduce an adaptive gamma correction (AGC) method to balance brightness equalization and increase the contrast of mammography images; (2) we use the ShuffleNet model to exchange information between different feature layers and extract the hidden high-level features of medical images; and (3) we use the transfer learning method to fine-tune the ShuffleNet model and obtain the optimal parameters.

Results:The proposed TumorDet model has shown that accuracy, sensitivity, and specificity reach 90.43%, 89.37%, and 87.81%, respectively. TumorDet performs well in the breast tumor detection task. In addition, we use the proposed TumorDet to conduct experiments on other tasks, such as forest fires, and the robustness of TumorDet is proved by experimental results.

Conclusion:TumorDet employs the ShuffleNet model to exchange information between different feature layers without increasing the number of network parameters and applies transfer learning method to further extract the basic features of medical images by fine-tuning. The model is beneficial for the localization and classification of breast tumors and also performs well in forest fire detection.

Current Bioinformatics. 2024;19(2):119-128
pages 119-128 views

Network Propagation-based Identification of Oligometastatic Biomarkers in Metastatic Colorectal Cancer

Jin Q., Yu K., Zhang X., Huo D., Zhang D., Liu L., Xie H., Liang B., Chen X.

Abstract

Background:The oligometastatic disease has been proposed as an intermediate state between primary tumor and systemically metastatic disease, which has great potential curable with locoregional therapies. However, since no biomarker for the identification of patients with true oligometastatic disease is clinically available, the diagnosis of oligometastatic disease remains controversial.

Objective:We aim to identify potential biomarkers of colorectal cancer patients with true oligometastatic states, who will benefit most from local therapy.

Methods:This study retrospectively analyzed the transcriptome profiles and clinical parameters of 307 metastatic colorectal cancer patients. A novel network propagation method and network-based strategy were combined to identify oligometastatic biomarkers to predict the prognoses of metastatic colorectal cancer patients.

Results:We defined two metastatic risk groups according to twelve oligometastatic biomarkers, which exhibit distinct prognoses, clinicopathological features, immunological characteristics, and biological mechanisms. The metastatic risk assessment model exhibited a more powerful capacity for survival prediction compared to traditional clinicopathological features. The low-MRS group was most consistent with an oligometastatic state, while the high-MRS might be a potential polymetastatic state, which leads to the divergence of their prognostic outcomes and response to treatments. We also identified 22 significant immune check genes between the high-MRS and low- MRS groups. The difference in molecular mechanism between the two metastatic risk groups was associated with focal adhesion, nucleocytoplasmic transport, Hippo, PI3K-Akt, TGF-β, and EMCreceptor interaction signaling pathways.

Conclusion:Our study provided a molecular definition of the oligometastatic state in colorectal cancer, which contributes to precise treatment decision-making for advanced patients.

Current Bioinformatics. 2024;19(2):129-143
pages 129-143 views

Thorough Assessment of Machine Learning Techniques for Predicting Protein-Nucleic Acid Binding Hot Spots

Zou X., Zhang C., Tang M., Deng L.

Abstract

Background:Proteins and nucleic acids are vital biomolecules that contribute significantly to biological life. The precise and efficient identification of hot spots at protein-nucleic acid interfaces is crucial for guiding drug development, advancing protein engineering, and exploring the underlying molecular recognition mechanisms. As experimental methods like alanine scanning mutagenesis prove to be time-consuming and expensive, a growing number of machine learning techniques are being employed to predict hot spots. However, the existing approach is distinguished by a lack of uniform standards, a scarcity of data, and a wide range of attributes. Currently, there is no comprehensive overview or evaluation of this field. As a result, providing a full overview and review is extremely helpful.

Methods:In this study, we present an overview of cutting-edge machine learning approaches utilized for hot spot prediction in protein-nucleic acid complexes. Additionally, we outline the feature categories currently in use, derived from relevant biological data sources, and assess conventional feature selection methods based on 600 extracted features. Simultaneously, we create two new benchmark datasets, PDHS87 and PRHS48, and develop distinct binary classification models based on these datasets to evaluate the advantages and disadvantages of various machine-learning techniques.

Results:Prediction of protein-nucleic acid interaction hotspots is a challenging task. The study demonstrates that structural neighborhood features play a crucial role in identifying hot spots. The prediction performance can be improved by choosing effective feature selection methods and machine learning methods. Among the existing prediction methods, XGBPRH has the best performance.

Conclusion:It is crucial to continue studying hot spot theories, discover new and effective features, add accurate experimental data, and utilize DNA/RNA information. Semi-supervised learning, transfer learning, and ensemble learning can optimize predictive ability. Combining computational docking with machine learning methods can potentially further improve predictive performance.

Current Bioinformatics. 2024;19(2):144-161
pages 144-161 views

iProm-Yeast: Prediction Tool for Yeast Promoters Based on ML Stacking

Shujaat M., Yoo S., Tayara H., Chong K.T.

Abstract

Background and Objective:Gene promoters play a crucial role in regulating gene transcription by serving as DNA regulatory elements near transcription start sites. Despite numerous approaches, including alignment signal and content-based methods for promoter prediction, accurately identifying promoters remains challenging due to the lack of explicit features in their sequences. Consequently, many machine learning and deep learning models for promoter identification have been presented, but the performance of these tools is not precise. Most recent investigations have concentrated on identifying sigma or plant promoters. While the accurate identification of Saccharomyces cerevisiae promoters remains an underexplored area. In this study, we introduced "iPromyeast", a method for identifying yeast promoters. Using genome sequences from the eukaryotic yeast Saccharomyces cerevisiae, we investigate vector encoding and promoter classification. Additionally, we developed a more difficult negative set by employing promoter sequences rather than nonpromoter regions of the genome. The newly developed negative reconstruction approach improves classification and minimizes the amount of false positive predictions.

Methods:To overcome the problems associated with promoter prediction, we investigate alternate vector encoding and feature extraction methodologies. Following that, these strategies are coupled with several machine learning algorithms and a 1-D convolutional neural network model. Our results show that the pseudo-dinucleotide composition is preferable for feature encoding and that the machine- learning stacking approach is excellent for accurate promoter categorization. Furthermore, we provide a negative reconstruction method that uses promoter sequences rather than non-promoter regions, resulting in higher classification performance and fewer false positive predictions.

Results:Based on the results of 5-fold cross-validation, the proposed predictor, iProm-Yeast, has a good potential for detecting Saccharomyces cerevisiae promoters. The accuracy (Acc) was 86.27%, the sensitivity (Sn) was 82.29%, the specificity (Sp) was 89.47%, the Matthews correlation coefficient (MCC) was 0.72, and the area under the receiver operating characteristic curve (AUROC) was 0.98. We also performed a cross-species analysis to determine the generalizability of iProm-Yeast across other species.

Conclusion:iProm-Yeast is a robust method for accurately identifying Saccharomyces cerevisiae promoters. With advanced vector encoding techniques and a negative reconstruction approach, it achieves improved classification accuracy and reduces false positive predictions. In addition, it offers researchers a reliable and precise webserver to study gene regulation in diverse organisms.

Current Bioinformatics. 2024;19(2):162-173
pages 162-173 views

In silico Study of Clinical Prognosis Associated MicroRNAs for Patients with Metastasis in Clear Cell Renal Carcinoma

Wijaya E.B., Mekala V.R., Zaenudin E., Ng K.

Abstract

Background:Metastasis involves multiple stages and various genetic and epigenetic alterations. MicroRNA has been investigated as a biomarker and prognostic tool in various cancer types and stages. Nevertheless, exploring the role of miRNA in kidney cancer remains a significant challenge, given the ability of a single miRNA to target multiple genes within biological networks and pathways.

Objective:This study aims to propose a computational research framework that hypothesizes that a set of miRNAs functions as key regulators in modulating gene expression networks of kidney cancer survival.

Methods:We retrieved the NGS data from the TCGA-KIRC extracted from UCSC Xena. A set of prognostic miRNAs was acquired through multiple Cox regression analyses. We adopted machine learning approaches to evaluate miRNA prognosis's classification performance between normal, primary (M0), and metastasis (M1) samples. The molecular mechanism between primary cancer and metastasis was investigated by identifying the regulatory networks of miRNA's target genes.

Results:A total of 14 miRNAs were identified as potential prognostic indicators. A combination of high-expression miRNAs was associated with survival probability. Machine learning achieved an average accuracy of 95% in distinguishing primary cancer from normal tissue and 79% in predicting the metastasis from primary tissue. Correlation analysis of miRNA prognostics with target genes unveiled regulatory network disparities between metastatic and primary tissues.

Conclusion:This study has identified 14 miRNAs that could potentially serve as vital biomarkers for diagnosing and prognosing ccRCC. Differential regulatory networks between metastatic and primary tissues in this study provide the molecular basis for assessment and therapeutic treatment for ccRCC patients.

Current Bioinformatics. 2024;19(2):174-192
pages 174-192 views