Abstract

Transcript levels are critical determinant of the proteome, and hence cellular function. Because the transcriptome is an outcome of the interactions between genes and their products, we reasoned it may be accurately represented by a subset of transcript abundances. By analyzing thousands of publicly available RNA -Seq datasets, we show that the transcriptome of Arabidopsis thaliana is highly compressible, and can be faithfully reconstructed using just the abundances of a small, machine-learned subset of 150 transcripts. These findings suggest natural improvements to both the time and cost of performing forward genetic screens and mapping eQTLs in naturally occurring populations. Understanding the rules by which the linear information encoded in the genome is read both in tissue and signalling contexts is a central aim of biology. Our transcriptome compression and prediction algorithms can facilitate dissection of this regulatory logic.