What Non Coding Genes Have Introns?

Question

[Mutated_Dater]

Hello

I'm a computer person learning somme biology. I've just wasted all afternoon trying to unravel the cause of some unexpected data and it turns out that non-coding RNA genes can have introns. I only know the basic biology dna-mrna-protein dogma. I am aware that certain rna genes exist too (for tRNA and rRNA and miRNA) but i had no idea there were other RNA genes that have introns. The gene in question was a LincRNA. In fact I'd never heard of this.

Are there any other RNA genes that have introns? Can pseudogenes have introns? Are there any other 'weird' genes i need to be aware of that are potential pitfalls when analysing genetic data? I get the feeling that the 80:20 rule comes up a lot in this field - you spend all of your time fixing the one or two odd scenarios that break your system.

thanks a lot

[David Quigley]

As a former software developer who has converted to genetics and bioinformatics, I think one of the hardest transitions to make is a way of thinking that is predicated on systems and rules that were devised by people for the goals of people. The systems we're working with are wet, messy, complicated, and follow rules that are often mysterious or hazy and incompletely revealed at best.

[Ketil]

Amen to that! It seems that any mechanism or phenomenon, no matter how contrieved or byzantine, actually exists, usually in some dark corner of biology.

Answer 1

[Larry_Parnell]

Yes, mRNAs of pseudogenes can be processed - i.e., have introns. In terms of "weird" genes, think about trans-splicing. There are excellent examples emerging from work in C. elegans, for example. Trans splicing can be thought of as a hybrid transcript - some exons from gene 1 and some from neighboring gene 2. LincRNA genes are a relatively new discovery and offer a great example of something different than "dogma."

Yes, these are biological phenomena and biological data - so they will likely break very strict code. That means that the 80:20 rule, or some variant of that, is certain to apply.

There are many single-exon genes, consisting of 5'-UTR, protein-coding segment, 3'-UTR all in one exon. Arabidopsis has a lot of these. Humans, too.

Answer 2

[User 59]

Xist is a completely awesome example of this. Not just a tiny, non-coding RNA with introns but a BIG one. Even more excellent in this tale is that it has an overlapping antisense, non-coding transcript called Tsix. Used to work in the same lab as some X-inactivation people, it's an interesting tale.

[Larry_Parnell]

HOTAIR is another.

[Larry_Parnell]

HYMAI is an interesting RNA gene. From RefSeq: HYMAI, hydatidiform mole associated and imprinted (non-protein coding), encodes a non-protein coding transcript and exhibits differential DNA methylation between the two parental alleles at an adjacent CpG island. HYMAI is expressed only from the paternal allele. It is believed to be one of the causative genes for transient neonatal diabetes mellitus.

Answer 3

[Jarretinha]

Complementing Larry answer, there still other classes of genes and problems. As a biologist that moved to bioinformatics, I can say for sure: Be prepared!!!

Gene processes/types not mentioned here but you must be aware:

• RNA editing (common in kinetoplastids, can be tought as a type of splicing)
• Alternative splicing (much more complicated in practice, lots of exceptions)
• Alternative (stop) codon translation (selenocysteine, pyrrolysine)
• Micro-exons (e. g. MEG genes in Schistosoma mansoni)
• Gene fusion (quite common in prokaryotes)

Just to mention, HOTAIR gene is a lincRNA that contains 6 exons and 5 introns. I've a spent a lot of time to find an example other than tRNA/rRNA genes. There's no mention to lincRNA gene structure in literature (ISI+SCOPUS+JSTOR)! But, when you look at the gene in NCBI you'll see the exon/intron structure. That's pretty amazing!!!

As a final point, even the exceptions have their own 80:20 rule. rRNA/tRNA genes in Prokarya have group I introns, Eukarya has group II. This changes everything. That's why I love biology and evolution!!!

So mutated_dater, welcome to the kingdom of surprises! You'll certainly find yours!

[Mutated_Dater]

thanks also for a great answer. I've read basic books but they don't tell you about things like trans splicing and micro exons and

Answer 4

[Gareth Palidwor]

tRNA introns are common (not sure but I don't think they're present in prokaryotes).

http://mips.helmholtz-muenchen.de/proj/yeast/reviews/intron/tRNA_com.html

The splicing mechanism is idiosyncratic, however.

[Gareth Palidwor]

the tRNA introns are tiny (<60 nt) so any potential encoded feature must be very small.

[Larry_Parnell]

Right. This brings up the topic of what is encoded in those introns. Could be a tRNA, microRNA, a protein-coding gene, anything really.

[Larry_Parnell]

To clarify - This brings up the topic of what is encoded in any intron be that from a protein-coding or RNA gene. Sure, if tRNA-encoding genes have small introns, then the possibilities are limited.

Answer 5

[Jorge Amigo]

the norm is that genes are divided in exons (coding regions) and introns (non coding regions), although I am not sure if there is any gene with a single exon only (in that case all gene would be coding DNA).

although such basic biology question should be covered by any general biology text book, you may quickly check it yourself on the second figure of the gene definition at wikipedia.

[Gareth Palidwor]

I hate to nitpick, but exons include UTRs as well as coding sequences (CDS). It's common to refer to use the term exon when referring to CDS, but the CDS is more precise.

http://en.wikipedia.org/wiki/Exon

[User 59]

That's not nit-picking, that's stating facts :)

[Mutated_Dater]

What code could i use to get the gene biotypes that have introns? I'm happier using the database directly than the api at present so an sql query would be great.

[Mutated_Dater]

i have read a basic biology textbook but this book never mentioned lincRNA which was what was causing the problem for me. in ensembl many 'genes' have one exon though they have their own definition of gene. By asking this question i was hoping to identify other less common 'genes'

[Gareth Palidwor]

I've hit the 80/20 rule a lot; it's a great, if painful, way to learn biology.

I doubt you'll find a list of all gene types and whether they have introns or not. What database are you using? I've learned a lot by exploring the data structure of Ensembl and looking at how the various gene types are structured. It would be simple enough to figure out what genes have introns with a little bit of code in the ensembl api...

[David Quigley]

The existence of lincrna (at least under that name) was only published in 2009 (Guttman Nature 2009), so it's not surprising that a general textbook would not mention them yet.

[Gareth Palidwor]

sorry, don't use the SQL directly at all anymore; API and SQL evolve in tandem and API hides the changes so I don't have to deal with them. I highly recommend the API over SQL.

The simplest (and nastiest) solution I can think of: for every gene count the number of exons. If any gene of a given biotype has more than one exon, then that biotype can have introns. Not a definitive answer for all organisms, but does tell you which biotypes can have introns and which don't for that organism...

Answer 6

[Peter]

Ribosomal RNAs frequently have introns.

[Larry_Parnell]

I know that several rRNAs can be transcribed in tandem and then processed to form mature species that are further modified at specific bases prior to incorporation into their respective ribosomal subunits.