The decreased expression of the CD3ζ chain in SLE T cells has been attributed to defects at multiple levels, including defective gene transcription [ 16], aberrant mRNA splicing [ 17], poor transcript stability of alternative splice variants [ 18], and increased protein degradation by caspase-mediated [ 19], ubiquitin-proteasome-mediated and lysosomal-mediated mechanisms [ 20].
Consequently, using just 1-kbp genomic regions as done previously [14], likely predict poor transcript models with high frequency.
For a more refined control of novel isoforms, or for organisms with poor transcript annotations, users are suggested to perform novel transcript assembly using tools like Cufflinks [22] and then load the resulting gene and transcript annotations (GTF) into PrimerSeq.
Whether low maternal transcript levels are the result of a lack or sub-optimal polyadenylation during Atlantic halibut oocyte maturation, leading to poor transcript translation and/or degradation in low quality early embryos, requires further investigation.
Consequently, our measurements of nucleotide-level changes were not dependent on the quality of the transcriptome, except that some regions of the gene may be missing due to poor transcript coverage.
For organisms with poor transcript annotations, we suggest researchers to use de novo RNA-seq transcript assembly tools, such as Cufflinks [22] and Scripture [23], to generate transcript annotations from their RNA-seq data, which can then be loaded into PrimerSeq for primer design.
The presence of non-orthologous genes in other species should be considered as a central resource to derive important information for the definition of gene models and structures, in particular for the newly assembled genomes, where the lack of a complete set of genes and poor transcript sequences information may restrict the commonly applied annotation procedures.
Clay et al. [ 14] looked at CpG islands upstream of GC-rich and GC-poor transcripts and found little correlation.
Overall, this study suggests that the short syncytial cycles in early Drosophila embryogenesis favor transcripts with a simple gene-architecture consisting of short, intron-poor transcripts (Fig. 2).
The conserved trend toward short, intron-poor transcripts among the first zygotically expressed genes extends to the mouse [ 44], even though the first cell cycles are longer than in fly, frog, or zebrafish embryos.
To examine a possible interrelation with the target prediction preference for long and GC-rich sequences, the 1,000 longest and GC-richest, as well as the 1,000 shortest and GC-poorest transcripts, were tested.
