#codons — Public Fediverse posts

When optimizing sequences for protein expression, we have to pay attention to preserving stop #codons in alternative frames, as demonstrated by a recent paper:

https://www.pnas.org/doi/full/10.1073/pnas.2606609123

This mirrors the relative high levels of AUGs in transcripts with alternative upstream #transcription start sites that we described in yeast. Such AUGs lead to short ORFs followed by long 3' UTR regions, perfect for degradation through #NMD:

https://elifesciences.org/articles/06722

#mRNA #RNA #science

Ribosome pausing, ribosome drop-off, an interesting study in the links between codons and translation in S. cerevisiae:

https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1012162

#science #ribosome #rna #codons

Short blog post about an R script that allows to optimize codons for reliable heterologous protein expression in various hosts. Requires a table of frequencies for allocation of codons to aminoacids for a given protein sequence.

https://cosminribo.eu/posts/2023-03-codons/

#translation #mRNA #Rstat #MolecularBiology #codons

The *split-apply-combine* method for data analysis is powerful. It is brilliantly used in the `dplyr` package and all over the `tidyverse` but base #R also works.

For example, for a table of codon frequencies with column "count" and column "Aa", relative frequencies of codons for each amino acid are calculated by:

relfreq <- function(v){v/sum(v)}

1. r1 <- split(mydf$count, f=list(mydf$Aa)

2. r2 <- lapply(r1, FUN=relfreq)

3. r3 <- unsplit(r2, f=list(mydf$Aa))

#DataAnalysis #codons