![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Over the break I read the first 5 chapters of Alberts et al, which I felt gave me a good grasp of biochemistry... but didn't really improve my understanding of genetics much. However, today, going to 3 classes, a lecture and a conversation, I have learned a lot.
A long long time ago (like 2008), I used to think that a gene was a letter in the {A,C,T,G} alphabet. But the scientific name for them is nucleotides (or bases).
A long time ago (like, yesterday), I used to think of genes as sequences of bases (or alternatively, of codons), demarcated by a start and a stop codon, sort of like quotation marks (I thought it would be trivial to have genes highlighted with the appropriate emacs mode).
This morning I learned about reading frames (set by the start codon), and mused that life might be funnier if the start and stop codons were inverse complements of each other, for this would mean that for every gene, there would be another gene that is its inverse complement and, if symmetry weren't broken, the same would happen to proteins too.
I can now see 2 ways in which this was wrong:
(a)
This afternoon, I learned that to be a gene you need to function as one, i.e. be eventually expressed into (one or more) proteins. I learned that you can fail to be a gene at many levels on the expression chain (e.g. the aminoacids won't fold into a stable configuration, mRNA won't get produced, etc), and these failures aren't always fully understood. This is why genes need to be annotated, and even then gene prediction isn't a solved problem (?).
Each level of failure corresponds to a notion of pseudogene. (Wikipedia says they are ex-genes, but wouldn't an accidental mutation into start and stop codons count as pseudogene too? How much function do you need to qualify as a pseudogene?)
(b)
Although both strands of DNA have a direction (from the 5' to the 3'), the 180-degree rotation isn't really a symmetry (even if the bases are). See: sense strand vs antisense strand. It's unclear what the physical basis of this is, and whether a given strand is sense or antisense is context-dependent rather than a constant for any given cell/organism/species.
I also learned that introns are more common the more complex the organism, enabling a larger splice-to-gene ratio (each splice produces a different mRNA). In humans, the average gene has 6 ways of splicing, with a huge amount of variation). Prokaryote genomes have no introns (?).
Misconception: I naively thought that stop codons signaled the beginning of introns, since they both signal the beginning of non-coding regions. I was immediately corrected.
A long long time ago (like 2008), I used to think that a gene was a letter in the {A,C,T,G} alphabet. But the scientific name for them is nucleotides (or bases).
A long time ago (like, yesterday), I used to think of genes as sequences of bases (or alternatively, of codons), demarcated by a start and a stop codon, sort of like quotation marks (I thought it would be trivial to have genes highlighted with the appropriate emacs mode).
This morning I learned about reading frames (set by the start codon), and mused that life might be funnier if the start and stop codons were inverse complements of each other, for this would mean that for every gene, there would be another gene that is its inverse complement and, if symmetry weren't broken, the same would happen to proteins too.
I can now see 2 ways in which this was wrong:
(a)
This afternoon, I learned that to be a gene you need to function as one, i.e. be eventually expressed into (one or more) proteins. I learned that you can fail to be a gene at many levels on the expression chain (e.g. the aminoacids won't fold into a stable configuration, mRNA won't get produced, etc), and these failures aren't always fully understood. This is why genes need to be annotated, and even then gene prediction isn't a solved problem (?).
Each level of failure corresponds to a notion of pseudogene. (Wikipedia says they are ex-genes, but wouldn't an accidental mutation into start and stop codons count as pseudogene too? How much function do you need to qualify as a pseudogene?)
(b)
Although both strands of DNA have a direction (from the 5' to the 3'), the 180-degree rotation isn't really a symmetry (even if the bases are). See: sense strand vs antisense strand. It's unclear what the physical basis of this is, and whether a given strand is sense or antisense is context-dependent rather than a constant for any given cell/organism/species.
I also learned that introns are more common the more complex the organism, enabling a larger splice-to-gene ratio (each splice produces a different mRNA). In humans, the average gene has 6 ways of splicing, with a huge amount of variation). Prokaryote genomes have no introns (?).
Misconception: I naively thought that stop codons signaled the beginning of introns, since they both signal the beginning of non-coding regions. I was immediately corrected.