Researchers have developed a new self-supervised learning method that can improve the training of genomic models with less labeled data. The method, called Self-GenomeNet, leverages reverse-complement sequences and effectively learns short- and long-term dependencies by predicting targets of different lengths. Self-GenomeNet outperforms other self-supervised methods in data-scarce genomic tasks and outperforms standard supervised training with ~10 times fewer labeled training data. Furthermore, the learned representations generalize well to new datasets and tasks. These findings suggest that Self-GenomeNet is well suited for large-scale, unlabeled genomic datasets and could substantially improve the performance of genomic models.
A study conducted bioinformatics analysis on GSE28829 and GSE43292 datasets to identify diagnostic markers and molecular mechanisms in atherosclerosis. Glutamine metabolism-associated genes were analyzed, resulting in 308 differentially expressed genes (DEGs). Weighted Gene Co-expression Network Analysis (WGCNA) and Protein–protein Interaction (PPI) network analysis revealed 27 hub genes, demonstrating high diagnostic values. Enrichment analyses showed associations […]
Next-generation protein sequencing (NGS) is a powerful new technology that can analyze proteins more comprehensively and in detail. This can be a valuable tool for researchers in various fields, including biology, medicine, and forensics.
A new machine learning approach developed by researchers at MIT and Harvard University could help scientists more efficiently design experiments to engineer cells into new states. This could have a significant impact on the development of new therapies for cancer, regenerative medicine, and other diseases. The new approach leverages the cause-and-effect relationship between factors in […]