Microsatellite instability, a genetic indication of DNA mismatch impairment, provides promising treatment options. Our study aimed to detect the mutation with whole-slide image (WSI) and discover the most effective pre-trained deep-learning model to sort diagnostic slides between high microsatellite instability (MSI-H) and microsatellite stable (MSS). WSI data retrieved from public dataset were processed for training and evaluating MSI categorization model. We detected MSI in slide levels for colorectal cancer (CRC), stomach adenocarcinoma (STAD), uterine corpus, and endometrial adenocarcinoma (UCEC). Models trained with a single tissue type were evaluated with the test dataset of corresponding tissue and subsequently with the test dataset of other types of tissue (cross-tissue evaluation). Finally, another model trained with multi-tissue types was built to predict the test dataset of individual tissue. Our models achieved AUC values of 0.93, 0.84, and 0.79 in TCGA-CRC, TCGA-STAD and TCGA-UCEC, respectively. We observed that a model trained on a corresponding tumor tissue demonstrates higher accuracy, particularly compared to those trained on other tumor tissues. In the combined model trained on multi-tissue, we observed diverse outcomes regarding which model was prioritized depending on the cancer type. These results demonstrate that models trained on multiple tissues have the potential to discern features that are generalizable across different types of cancer.
Background: Recent deep sequencing technologies have proven to be valuable resources to gain insights into the expression profiles of diverse tRNAs. However, despite these technologies, the association of tRNAs with diverse diseases has not been explored in depth because analytical tools are lacking. Results: We developed a user-friendly tool, tRNA Expression Analysis Software Utilizing R for Easy use (tReasure), to analyze differentially expressed tRNAs (DEtRNAs) from deep sequencing data of small RNAs using R packages. tReasure can quantify individual mature tRNAs, isodecoders, and isoacceptors. By adopting stringent mapping strategies, tReasure supports the precise measurement of mature tRNA read counts. The whole analysis workflow for determining DEtRNAs (uploading FASTQ files, removing adapter sequences and poor-quality reads, mapping and quantifying tRNAs, filtering out low count tRNAs, determining DEtRNAs, and visualizing statistical analysis) can be performed with the tReasure package. Conclusions: tReasure is an open-source software available for download at https://treasure.pmrc.re.kr and will be indispensable for users who have little experience with command-line software to explore the biological implication of tRNA expression.
Transfer RNA (tRNA), a key component of the translation machinery, plays critical roles in stress conditions and various diseases. While knowledge regarding the importance of tRNA function is increasing, its biological roles are still not well understood. There is currently no comprehensive database or web server providing the expression landscape of tRNAs across a variety of human tissues and diseases. Here, we constructed a user-friendly and interactive database, DBtRend, which provides a profile of mature tRNA expression across various biological conditions by reanalyzing the small RNA or microRNA sequencing data from the Cancer Genome Atlas (TCGA) and NCBI’s Gene Expression Omnibus (GEO) in humans. Users can explore not only the expression values of mature individual tRNAs in the human genome, but also those of isodecoders and isoacceptors based on our specific pipelines. DBtRend provides the expressed patterns of tRNAs, the differentially expressed tRNAs in different biological conditions, and the information of samples or patients, tissue types, and molecular subtype of cancers. The database is expected to help researchers interested in functional discoveries of tRNAs.