LegNet: a best-in-class deep learning model for short DNA regulatory regions.

MOTIVATION: The increasing volume of data from high-throughput experiments including parallel reporter assays facilitates the development of complex deep-learning approaches for modeling DNA regulatory grammar. RESULTS: Here, we introduce LegNet, an EfficientNetV2-inspired convolutional network for modeling short gene regulatory regions. By approaching the sequence-to-expression regression problem as a soft classification task, LegNet secured first place for the autosome.org team in the DREAM 2022 challenge of predicting gene expression from gigantic parallel reporter assays. Using published data, here, we demonstrate that LegNet outperforms existing models and accurately predicts gene expression per se as well as the effects of single-nucleotide variants. Furthermore, we show how LegNet can be used in a diffusion network manner for the rational design of promoter sequences yielding the desired expression level. AVAILABILITY AND IMPLEMENTATION: https://github.com/autosome-ru/LegNet. The GitHub repository includes Jupyter Notebook tutorials and Python scripts under the MIT license to reproduce the results presented in the study.

Tyrosol induces multiple drug resistance in yeast Saccharomyces cerevisiae.

In yeast, multiple (pleiotropic) drug resistance (MDR) transporters efflux xenobiotics from the cytoplasm to the environment. Additionally, upon the accumulation of xenobiotics in the cells, MDR genes are induced. At the same time, fungal cells can produce secondary metabolites with physico-chemical properties similar to MDR transporter substrates. Nitrogen limitation in yeast Saccharomyces cerevisiae leads to the accumulation of phenylethanol, tryptophol, and tyrosol, which are products of aromatic amino acid catabolism. In this study, we investigated whether these compounds could induce or inhibit MDR in yeast. Double deletion of PDR1 and PDR3 genes, which are transcription factors that upregulate the expression of PDR genes, reduced yeast resistance to high concentrations of tyrosol (4-6 g/L) but not to the other two tested aromatic alcohols. PDR5 gene, but not other tested MDR transporter genes (SNQ2, YOR1, PDR10, PDR15) contributed to yeast resistance to tyrosol. Tyrosol inhibited the efflux of rhodamine 6G (R6G), a substrate for MDR transporters. However, preincubating yeast cells with tyrosol induced MDR, as evidenced by increased Pdr5-GFP levels and reduced yeast ability to accumulate Nile red, another fluorescent MDR-transporter substrate. Moreover, tyrosol inhibited the cytostatic effect of clotrimazole, the azole antifungal. Our results demonstrate that a natural secondary metabolite can modulate yeast MDR. We speculate that intermediates of aromatic amino acid metabolites coordinate cell metabolism and defense mechanisms against xenobiotics.

ORFans in Mitochondrial Genomes of Marine Polychaete Polydora.

Most characterized metazoan mitochondrial genomes are compact and encode a small set of proteins that are essential for oxidative phosphorylation, as well as rRNA and tRNA for their expression. However, in rare cases, invertebrate taxa have additional open reading frames (ORFs) in their mtDNA sequences. Here, we sequenced and analyzed the mitochondrial genome of a polychaete worm, Polydora cf. ciliata, part of whose life cycle takes place in low-oxygen conditions. In the mitogenome, we found three "ORFan" regions (544, 1,060, and 427 bp) that have no resemblance to any standard metazoan mtDNA gene but lack stop codons in one of the reading frames. Similar regions are found in the mitochondrial genomes of three other Polydora species and Bocardiella hamata. All five species share the same gene order in their mitogenomes, which differ from that of other known Spionidae mitogenomes. By analyzing the ORFan sequences, we found that they are under purifying selection pressure and contain conservative regions. The codon adaptation indices (CAIs) of the ORFan genes were in the same range of values as the CAI of conventional protein-coding genes in corresponding mitochondrial genomes. The analysis of the P. cf. ciliata mitochondrial transcriptome showed that ORFan-544, ORFan-427, and a portion of the ORFan-1060 are transcribed. Together, this suggests that ORFan-544 and ORFan-427 encode functional proteins. It is likely that the ORFans originated when the Polydora/Bocardiella species complex separated from the rest of the Spionidae, and this event coincided with massive gene rearrangements in their mitochondrial genomes and tRNA-Met duplication.