The yeast Mip6 multi-omics dataset was created with a dual goal. On the one hand, to study the crosstalk between epigenetic, transcriptional and metabolic regulation in response to environmental changes and in the presence of the RNA binding protein Mip6. On the other hand, to facilitate the integration and analysis of data using covariance-based methods.
Our experimental design was based on a strategy to obtain: i) all multi-omic datasets from the same culture ii) four biological replicates. This strategy was key to ensure the reproducibility as well as for biological consistency. Most of the studies use two or three replicates tend to obtain each omic from a different sample. This is due to the difficulties on accessing to biological material or the cost of making more replicates. We scaled up the volume of our starting culture to obtain sufficient material for each omic. Moreover, we collected more material for ChIP-seq, since the yield of chromatin immunoprecipitation with antibodies against specific histone modifications is not always optimal. This allowed us to repeat some ChIP experiments from the exact same cells that were stored for later use. Although our intention was to obtain all samples at the same time, owing to sample management limitations, these 4 replicates were obtained on two different days. However, this enabled the analysis of different batches, the optimization of batch effect analyses and accounting for the slight batch effect for the day of culture growth.
Our interest in this study was to understand how different molecular layers that regulate gene expression under heat stress in yeast and its dependence on Mip6, a Mex67-interacting factor that contributes to maintaining low levels of specific stress-dependent mRNAs (Martín-Expósito et al., EMBO Reports 2019). We have learnt that H4K12 acetylation responds more rapidly to heat stress than gene expression. Our analysis showed that the gene expression response measured by RNA-seq has a larger dynamic range than H4K12 acetylation measured by ChIP-seq. Both signals are coordinated but their magnitude of change differs, with RNA-seq data manifesting a larger dynamic range.
We could also show for the first time that Mip6 is required for the correct expression of trehalose metabolism genes, which correlates with an accumulation of this metabolite in our metabolomics dataset. We plan for further studies on the biological relevance of Mip6 in the stress response, by thoroughly analyzing this multi-omic dataset and comparing it with accessible data from other teams.
We hope that our data can be useful to other researchers in the field and that our strategy will be implemented when planning their multi-omics experiments.