Researchers from the Cold Spring Harbor Laboratory and the University of British Columbia present a differential gene expression predictor developed with array- or RNA sequencing-based expression data from 635 available studies spanning more than 27,000 samples. The authors compiled and re-analyzed the data in a University of British Columbia Gemma database, demonstrating that "a gene's prior probability of differential expression allows for accurate prediction of [differential expression] hit lists, regardless of the biological question." The team subsequently applied the predictor, known as the DE prior, to interpret expression markers associated with breast cancer subtypes, pancreatic islet cells assessed with single-cell RNA-seq, and meta-analyses focused on lung adenocarcinoma or kidney transplant rejection samples.
A team from Cornell University and the Chinese Academy of Agricultural Sciences explore deep machine learning methods for predicting transcript abundance based on DNA sequences with the help of evolutionary insights. The researchers focused on two deep learning models that considered evolutionary relationships, including a gene-family guided splitting method and a method that involves comparisons between orthologous gene pairs. "The methods are tested and applied within the context of predicting [messenger RNA] expression levels from whole-genome DNA sequence data," they find, "and are applicable across biological organisms."
Investigators at the University of Amsterdam and elsewhere consider clonal evolution dynamics over space and time in colorectal cancer (CRC). Using a lentiviral gene ontology-based vectors, the researchers labeled primary colon cancer cell lines and human CRC xenograft models grown in mice with red, green, and blue markers to follow tumor growth. In these experiments and in computational models, they saw clonal growth differences related to a clone's location in the tumor. "Our findings suggest that either microenvironmental signals on the tumor border or differences in physical properties within the tumor, are major contributors to explain heterogeneous clonal expansion," the authors report, noting that the study "provides further insights into the dynamics of solid tumor growth and progression."