Such compensatory changes that underlie the91expression phenotype are expected to evolve under stabilizing selection and often result in92developmental system drift, a phenomenon whereby gene expression phenotype is maintained93between species despite molecular evolution of gene regulatory sequences
major theory idea here is that the same phenotype can be produced my multiple different combinations of regulatory backgrounds
Despite the considerable efforts made in studying the62genetic basis of post-zygotic reproductive isolation for almost a century, and the aforementioned63broad trends, a consensus regarding the relative importance of specific causal mechanisms across64different taxa has not yet been reached.
BIG question: how can we characterize the relative effects of different mechanisms on the genetic basis of post-zygotic reproductive isolation?
Figure 3.
i hate these figures
Approximately 15% of bases coveredby exons of gene models annotated in WormBase are detected in all25 cell types profiled (Fig. 4C). A largerfraction of bases derived from gene mod-els (about 75%), however, is expressed inat least one, but not all, of the cell types,and 11% is detected in no more than twocell types. For stages, we observed that29% of bases from exons of annotatedgene models is detected throughout de-velopment versus 8% expressed in nomore than one developmental period(Supplemental Fig. S17).
more total ubiquitous genes than specific to any total time point or tissue
These lists of selectively enriched genescomprised about 20%–57% of all genesenriched in the corresponding tissue orcell type. Eighty-two percent of all genesselectively enriched in any cell type ortissue are specific to only one or twosamples.
seems like most (?) transcribed genes for any given tissue type are specifically enriched there
The information theoretic conceptof Shannon entropy
what the fuck does this even mean lmao
Figure 2.
who put this on a separate page :/
Lists of cell or tissue-enriched tran-scripts are included as supplemental data(http://www.vanderbilt.edu/wormdoc/wormmap/) and should be useful foridentifying genes with key roles in thecorresponding cell type (see also Sup-plemental Results SR2).
useful tool for future research
On the basis of these results, we concludethat transcripts for a majority of C. elegans genes are regulated toachieve different levels of expression during development andbetween specific types of cells.
shocker
snoRNAs
Small nucleolar RNAs (snoRNAs) are a class of non-coding small RNA molecules in the nucleolus that primarily guide chemical modifications of other RNAs, mainly ribosomal RNAs, transfer RNAs and small nuclear RNAs. - Wikipedia
ncRNAs
non-coding
Taken to-gether, most gene models (about 90%) were supported by nrTARs,whereas a substantial fraction of nrTARs could not simply be at-tributed to known transcripts
not every transcriptionally active region mapped to a gene that was already known
For the segmentation of hybridization signals intointergenic regions, exons, and introns, we used a method calledmSTAD
compared this to another tiling method and already annotated genes to make sure that it works
Thus, a global analysis of our tilingarray results suggests that cell-specific profiling preserves overallpatterns of temporally regulated gene expression
global expression changes over time (i.e. embryo cell likely has more similar expression to total embryonic transcriptome than that same cell in an adult)
Figure 1
shows various ways to collect RNA data specific or non-specific to a cell type or time period
Thus, the anatomy and development of theanimal is defined at the resolution of the single cell, but a compa-rably precise atlas of gene expression is not currently available.
big Q: how does C elegans' gene expression vary across development/anatomy, can that be mapped?
Among these, expressionof genes involved in reproductive processes (e.g. regula-tion of meiotic cell cycle, reciprocal meiotic recombin-ation, meiotic chromosome segregation) tended tocovary positively with fecundity. By contrast, expressionof genes involved in behavior and neuromuscular func-tion (e.g. chemical synaptic transmission, axon guid-ance, neuropeptide signaling, regulation of musclecontraction) tended to covary negatively with fecundity.
using energy on anything other than fecundity = less fecundity
By contrast, oldergenes show stronger covariance with fecundity, consist-ent with their deeper integration into conserved physio-logical and regulatory network
tracks with the rest of the findings so far -> > necessity = > covariance
Tissue-specific Heterogeneity of GeneExpression-fitness Covariance
tldr: covariance is dependent on tissue location
Genesmarked by increasingly accessible chromatin exhibitedlower dN/dS values
makes sense, more accessible = more important = more dangerous to vary
Together, these patterns areconsistent with a polygenic architecture of fecundity(Zhang et al. 2021), in which gene expression-fitness co-variance is distributed across many highly regulated tar-get genes, rather than concentrated among a fewhighly connected TFs (Liu et al. 2019)
gene expression/fitness covariance depends more on lots of genes regulated by many TFs vs on a few TFs that regulate lots of genes
Thus, genes with more broadly accessibleregulatory landscapes show stronger covariance be-tween transcript abundance and fecundity of wormsin the novel laboratory environment.
more accessible = more complex = more covariance between transcript abundance and fecundity
why? idk
they do not drive genome-widepatterns, further supporting the robustness of our globalestimates. We therefore proceed with analyses of globalpatterns in S, while controlling for genomic context in tar-geted downstream analyses where relevant.
selective sweeps and HDRs don't play a major role in genome-wide associations between fitness and transcript level
selection may targetcoordinated expression patterns of multiple genes ra-ther than abundance of individual transcripts, consist-ent with the omnigenic model of expression evolution
targets GRN not one transcript
suggestingthat in the new laboratory environment, expression vari-ance tends to be disadvantageous, consistent with priorfindings that suggest pervasive weak stabilizing selec-tion influences the evolution of genome-wide gene ex-pression patterns in C. elegans
variance bad bc lack of environmental adversity
S, representing directional se-lection (whether transcript abundance covaries positive-ly or negatively with fitness)
does better fitness equal more transcript?
C, representingstabilizing or diversifying selection (selection for oragainst population-level variance in gene expression, re-spectively)
greater or less variance
However, a key challenge in interpreting trait-fitness as-sociations in C. elegans is its strong population structure dri-ven by hermaphroditic reproduction.
issue with looking at this in caenorhabditis is the strong prevalence of linkage/linked selection - skews associations between genes and phenotype
Yet, results vary across sys-tems. For example, population-scale variation in transcriptabundance displayed negative covariance with fitness formost genes in rice cultivated under traditional paddy fieldconditions (Groen et al. 2020; Gupta et al. 2025), whereasopposite trends were observed in field-sampled salmonidfish (Ahmad et al. 2021). These mixed patterns challengethe expectation from comparative studies that stabilizingselection should constrain gene expression divergence(Bedford and Hartl 2009; Kalinka et al. 2010; Chen et al.2019; El Taher et al. 2021), raising critical questions abouthow GRNs diverge across scales and species.
previous research is inconclusive/conflicting
In doing so, it allowsus to ask whether the same principles that govern GRN di-vergence over long timescales also manifest in microevolu-tionary patterns of gene expression-fitness covariancewithin a single generation.
also looking at similarities/differences between long-term evolution and change within a generation
exhibit weak coupling between ratesof sequence and expression evolution
sequence + expression evolve independently?
omnigenic model” proposesthat most gene expression differences in populationsare influenced by trans-regulatory variants scatteredthroughout the genome, reinforcing the notion thatGRNs function as distributed systems with many indirectinfluences
variance from trans reg elements and not the sequence itself
Genes occupying central or highly con-nected positions within GRNs tend to show greaterpleiotropy and are consequently subject to strongerconstraints (Josephs et al. 2017). Younger genes, onthe other hand, are often enriched at the periphery ofGRNs, where constraints are relaxed, enabling expres-sion divergence (Capra et al. 2010; Wei et al. 2016;Mähler et al. 2017; Defoort et al. 2018)
< total functions = easier to change
Aspopulations diverge, changes in gene expression withinGRNs can arise through both adaptive processes andneutral genetic drift; yet, the relative importance ofeach remains unresolved
big q: relative importance of drift vs adaptation on evolution of GRNs
Aspopulations diverge, changes in gene expression withinGRNs can arise through both adaptive processes andneutral genetic drift; yet, the relative importance ofeach remains unresolved
big question: what is the relative importance of drift vs adaptation on changes in the GRN