A random forest classifier for protein-protein docking models

An Ensemble Classifiers for Improved Prediction of Native-Non-Native Protein-Protein Interaction

In this study, we present an innovative approach to improve the prediction of protein-protein interactions (PPIs) through the utilization of an ensemble classifier, specifically focusing on distinguishing between native and non-native interactions. Leveraging the strengths of various base models, including random forest, gradient boosting, extreme gradient boosting, and light gradient boosting, our ensemble classifier integrates these diverse predictions using a logistic regression meta-classifier. Our model was evaluated using a comprehensive dataset generated from molecular dynamics simulations. While the gains in AUC and other metrics might seem modest, they contribute to a model that is more robust, consistent, and adaptable. To assess the effectiveness of various approaches, we compared the performance of logistic regression to four baseline models. Our results indicate that logistic regression consistently underperforms across all evaluated metrics. This suggests that it may not be well-suited to capture the complex relationships within this dataset. Tree-based models, on the other hand, appear to be more effective for problems involving molecular dynamics simulations. Extreme gradient boosting (XGBoost) and light gradient boosting (LightGBM) are optimized for performance and speed, handling datasets effectively and incorporating regularizations to avoid over-fitting. Our findings indicate that the ensemble method enhances the predictive capability of PPIs, offering a promising tool for computational biology and drug discovery by accurately identifying potential interaction sites and facilitating the understanding of complex protein functions within biological systems.

A Survey of Deep Learning Methods for Estimating the Accuracy of Protein Quaternary Structure Models

The quality prediction of quaternary structure models of a protein complex, in the absence of its true structure, is known as the Estimation of Model Accuracy (EMA). EMA is useful for ranking predicted protein complex structures and using them appropriately in biomedical research, such as protein-protein interaction studies, protein design, and drug discovery. With the advent of more accurate protein complex (multimer) prediction tools, such as AlphaFold2-Multimer and ESMFold, the estimation of the accuracy of protein complex structures has attracted increasing attention. Many deep learning methods have been developed to tackle this problem; however, there is a noticeable absence of a comprehensive overview of these methods to facilitate future development. Addressing this gap, we present a review of deep learning EMA methods for protein complex structures developed in the past several years, analyzing their methodologies, data and feature construction. We also provide a prospective summary of some potential new developments for further improving the accuracy of the EMA methods.

Anticancer Peptides Derived from Aldolase A and Induced Tumor-Suppressing Cells Inhibit Pancreatic Ductal Adenocarcinoma Cells

PDAC (pancreatic ductal adenocarcinoma) is a highly aggressive malignant tumor. We have previously developed induced tumor-suppressing cells (iTSCs) that secrete a group of tumor-suppressing proteins. Here, we examined a unique procedure to identify anticancer peptides (ACPs), using trypsin-digested iTSCs-derived protein fragments. Among the 10 ACP candidates, P04 (IGEHTPSALAIMENANVLAR) presented the most efficient anti-PDAC activities. P04 was derived from aldolase A (ALDOA), a glycolytic enzyme. Extracellular ALDOA, as well as P04, was predicted to interact with epidermal growth factor receptor (EGFR), and P04 downregulated oncoproteins such as Snail and Src. Importantly, P04 has no inhibitory effect on mesenchymal stem cells (MSCs). We also generated iTSCs by overexpressing ALDOA in MSCs and peripheral blood mononuclear cells (PBMCs). iTSC-derived conditioned medium (CM) inhibited the progression of PDAC cells as well as PDAC tissue fragments. The inhibitory effect of P04 was additive to that of CM and chemotherapeutic drugs such as 5-Flu and gemcitabine. Notably, applying mechanical vibration to PBMCs elevated ALDOA and converted PBMCs into iTSCs. Collectively, this study presented a unique procedure for selecting anticancer P04 from ALDOA in an iTSCs-derived proteome for the treatment of PDAC.

Anticancer peptides from induced tumor-suppressing cells for inhibiting osteosarcoma cells

Osteosarcoma (OS) is the most frequent primary bone cancer, which is mainly suffered by children and young adults. While the current surgical treatment combined with chemotherapy is effective for the early stage of OS, advanced OS preferentially metastasizes to the lung and is difficult to treat. Here, we examined the efficacy of ten anti-OS peptide candidates from a trypsin-digested conditioned medium that was derived from the secretome of induced tumor-suppressing cells (iTSCs). Using OS cell lines, the antitumor capabilities of the peptide candidates were evaluated by assaying the alterations in metabolic activities, proliferation, motility, and invasion of OS cells. Among ten candidates, peptide P05 (ADDGRPFPQVIK), a fragment of aldolase A (ALDOA), presented the most potent OS-suppressing capabilities. Its efficacy was additive with standard-of-care chemotherapeutic agents such as cisplatin and doxorubicin, and it downregulated oncoproteins such as epidermal growth factor receptor (EGFR), Snail, and Src in OS cells. Interestingly, P05 did not present inhibitory effects on non-OS skeletal cells such as mesenchymal stem cells and osteoblast cells. Collectively, this study demonstrated that iTSC-derived secretomes may provide a source for identifying anticancer peptides, and P05 may warrant further evaluations for the treatment of OS.

Discriminating physiological from non-physiological interfaces in structures of protein complexes: A community-wide study

Reliably scoring and ranking candidate models of protein complexes and assigning their oligomeric state from the structure of the crystal lattice represent outstanding challenges. A community-wide effort was launched to tackle these challenges. The latest resources on protein complexes and interfaces were exploited to derive a benchmark dataset consisting of 1677 homodimer protein crystal structures, including a balanced mix of physiological and non-physiological complexes. The non-physiological complexes in the benchmark were selected to bury a similar or larger interface area than their physiological counterparts, making it more difficult for scoring functions to differentiate between them. Next, 252 functions for scoring protein-protein interfaces previously developed by 13 groups were collected and evaluated for their ability to discriminate between physiological and non-physiological complexes. A simple consensus score generated using the best performing score of each of the 13 groups, and a cross-validated Random Forest (RF) classifier were created. Both approaches showed excellent performance, with an area under the Receiver Operating Characteristic (ROC) curve of 0.93 and 0.94, respectively, outperforming individual scores developed by different groups. Additionally, AlphaFold2 engines recalled the physiological dimers with significantly higher accuracy than the non-physiological set, lending support to the reliability of our benchmark dataset annotations. Optimizing the combined power of interface scoring functions and evaluating it on challenging benchmark datasets appears to be a promising strategy.

Exploration of m6A methylation regulators as epigenetic targets for immunotherapy in advanced sepsis

BACKGROUND

This study aims to deeply explore the relationship between m⁶A methylation modification and peripheral immune cells in patients with advanced sepsis and mine potential epigenetic therapeutic targets by analyzing the differential expression patterns of m⁶A-related genes in healthy subjects and advanced sepsis patients.

METHODS

A single cell expression dataset of peripheral immune cells containing blood samples from 4 patients with advanced sepsis and 5 healthy subjects was obtained from the gene expression comprehensive database (GSE175453). Differential expression analysis and cluster analysis were performed on 21 m⁶A-related genes. The characteristic gene was identified based on random forest  algorithm, and the correlation between the characteristic gene METTL16 and 23 immune cells in patients with advanced sepsis was evaluated using single-sample gene set enrichment analysis.

RESULTS

IGFBP1, IGFBP2, IGF2BP1, and WTAP were highly expressed in patients with advanced sepsis and m⁶A cluster B. IGFBP1, IGFBP2, and IGF2BP1 were positively correlated with Th17 helper T cells. The characteristic gene METTL16 exhibited a significant positive correlation with the proportion of various immune cells.

CONCLUSION

IGFBP1, IGFBP2, IGF2BP1, WTAP, and METTL16 may accelerate the development of advanced sepsis by regulating m⁶A methylation modification and promoting immune cell infiltration. The discovery of these characteristic genes related to advanced sepsis provides potential therapeutic targets for the diagnosis and treatment of sepsis.

Improving classification of correct and incorrect protein-protein docking models by augmenting the training set

Motivation

Protein-protein interactions drive many relevant biological events, such as infection, replication and recognition. To control or engineer such events, we need to access the molecular details of the interaction provided by experimental 3D structures. However, such experiments take time and are expensive; moreover, the current technology cannot keep up with the high discovery rate of new interactions. Computational modeling, like protein-protein docking, can help to fill this gap by generating docking poses. Protein-protein docking generally consists of two parts, sampling and scoring. The sampling is an exhaustive search of the tridimensional space. The caveat of the sampling is that it generates a large number of incorrect poses, producing a highly unbalanced dataset. This limits the utility of the data to train machine learning classifiers.

Results

Using weak supervision, we developed a data augmentation method that we named hAIkal. Using hAIkal, we increased the labeled training data to train several algorithms. We trained and obtained different classifiers; the best classifier has 81% accuracy and 0.51 Matthews' correlation coefficient on the test set, surpassing the state-of-the-art scoring functions.

Availability and implementation

Docking models from Benchmark 5 are available at https://doi.org/10.5281/zenodo.4012018. Processed tabular data are available at https://repository.kaust.edu.sa/handle/10754/666961. Google colab is available at https://colab.research.google.com/drive/1vbVrJcQSf6\_C3jOAmZzgQbTpuJ5zC1RP?usp=sharing.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.