Skip to main content
Premium Trial:

Request an Annual Quote

Nature Papers on Predicting Single-Cell Expression From Bulk Data, Pyrimidine Nucleobases Uncovered in Meteorites, More

A statistical method for predicting cellular composition and gene expression in individual cell types from bulk RNA sequencing data is presented in Nature Cancer this week. Called BayesPrism and developed by researchers from Cornell University and the Sloan Kettering Institute, the Bayesian model jointly infers the posterior distribution of cell type fractions and gene expression from bulk RNA-seq datasets using patient-derived, single-cell RNA-seq reference as prior information. "By explicitly modeling and marginalizing out the differences in gene expression between single-cell reference and bulk data, BayesPrism substantially outperforms leading methods in the inference of cell type fractions in both tumor and nontumor settings," the scientists write. The model is demonstrated using a large dataset of bulk RNA-seq and scRNA-seq samples in glioblastoma, head and neck squamous cell carcinoma, and skin cutaneous melanoma.

The pyrimidine nucleobases that are required for the formation of DNA and RNA have been found in carbon-rich meteorites, suggesting that the emergence of genetic properties for the earliest life on Earth began in outer space, according to a new study in Nature Communications. To date, only purine nucleobases such as guanine and adenine have been detected in meteorites, but lab experiments have suggested that pyrimidines — which include cytosine, uracil, and thymine — can be produced from chemical precursors found in these interstellar objects. To investigate, a multi-institute team of Japanese scientists used state-of-the-art analytical techniques optimized for small-scale quantification of nucleobases to examine three carbonaceous meteorites. In addition to previously detected purine nucleobases in meteorites, they identify various pyrimidine nucleobases such as cytosine, uracil, and thymine, as well as their structural isomers. "Given the similarity in the molecular distribution of pyrimidines in meteorites and those in photon-processed interstellar ice analogues, some of these derivatives could have been generated by photochemical reactions prevailing in the interstellar medium and later incorporated into asteroids during solar system formation," they write. As a result, meteoric nucleobases may have been the building blocks of DNA and RNA on the early Earth.

A new tool for RNA counting in single-cell RNA sequencing experiments is described in Nature Methods this week. Within scRNA-seq, counting strategies are used to mitigate the overcounting of amplicons derived from each RNA or DNA molecule, yet no experimental strategy to evaluate counting performance exists. To address this, Karolinska Institute scientists developed what they term molecular spikes: RNA spike-ins containing built-in unique molecular identifiers that can be used to identify critical experimental and computational conditions for accurate RNA counting in scRNA-seq. Using molecular spikes, they uncovered impaired RNA counting in popular scRNA-seq methods. The researchers also show how molecular spikes can be used to improve estimates of total endogenous RNA amounts in cells and rescue faulty experiments.