Specialized RNA sequencing
Full-length cDNA from oligo-dT priming and template switching is amplified, made into circular templates and read by circular consensus, so each accurate long read is one complete transcript from poly(A) tail to 5′ end.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
Iso-Seq
Full-length cDNA from oligo-dT priming and template switching is amplified, made into circular templates and read by circular consensus, so each accurate long read is one complete transcript from poly(A) tail to 5′ end.
How it differs from the standard method
- Start from intact RNA (RIN ≥ 8); degradation creates false 5′ ends.
- Use the fewest PCR cycles that give enough cDNA, and consider adding back a long size-selected fraction.
- Keep only full-length non-chimeric reads before clustering into isoforms.
- Concatenation (array) approaches put several cDNAs in one read for more reads per run.
1Questions that short-read counts cannot answer
Standard RNA-seq breaks transcripts into short fragments and counts them per gene. That loses three kinds of information.
Whole transcripts as cDNA
Full-length cDNA is made by oligo-dT priming and template switching, then each molecule is read end to end as an accurate circular-consensus long read.
Native RNA through a pore
A motor protein feeds the RNA through a nanopore from its poly(A) tail; the current reports the bases, modified bases such as m6A, and the tail length.
Where ribosomes sit
Nuclease digestion leaves ~30 nt footprints under each ribosome; their positions show which reading frames are translated and how strongly.
2How each one goes wrong
- Iso-Seq: degraded RNA gives cDNAs that start downstream of the real 5′ end and look like new start sites. Short molecules amplify and load preferentially, so long isoforms are under-counted. Over-cycled PCR makes chimeras that look like fusion transcripts.
- Direct RNA: reads run from the 3′ end and stop where the RNA is broken, so RNase damage costs coverage far from the tail. Throughput is modest and needs hundreds of nanograms of poly(A) RNA. Modification calls have a false-positive rate that only an unmodified control can measure.
- Ribo-seq: under-digestion leaves long, ragged footprints with no codon resolution, and over-digestion damages ribosomes. rRNA fragments of footprint size dominate unless depleted. Warm cycloheximide pre-treatment piles ribosomes on start codons. Without matched RNA-seq, footprints cannot be turned into translation efficiency.
3Reading the results
Long-read isoforms are classified against the annotation: full splice matches, incomplete matches (often truncations), and novel combinations or junctions. Novel isoforms need support from several full-length reads and, ideally, independent evidence, since artefacts are common.
Key terms
Practise these as flashcards →Reverse transcriptase adds untemplated cytosines at the 5′ end of the RNA, where an oligo anneals so the full cDNA carries an adapter at both ends.
Full-length non-chimeric: a consensus read with both cDNA primers and a poly(A) tail, and no internal primer.
Full splice match and incomplete splice match: categories that compare a long-read isoform with the annotated ones.
N6-methyladenosine, the most common internal mRNA modification, usually at DRACH sequence motifs.
The fraction of molecules modified at a given site.
The ~28–30 nt piece of mRNA protected from nuclease by a ribosome.
The distance from a footprint's 5′ end to the codon in the ribosome's P site, used to place each footprint on a codon.
Footprint positions concentrated in one of the three frames along a coding sequence, the signature of translating ribosomes.
Footprint density on a coding sequence divided by its mRNA abundance.
When to use it
- Cataloguing full-length isoforms, alternative start and end sites, or fusion transcripts without inference from short reads.
- Measuring RNA modifications such as m6A, or poly(A) tail length, on native molecules.
- Asking whether a change in protein comes from translation rather than mRNA level, finding translated upstream ORFs, or mapping translation start sites.
Compared with alternatives
Cheaper and far deeper for gene-level counting and differential expression; isoforms are inferred rather than observed.
Covers the platforms and their DNA uses; this page is about the RNA library choices on top of them.
Maps where RNA-binding proteins (including m6A readers or ribosome-associated factors) sit on RNA by crosslinking and immunoprecipitation.
Measures the protein products directly; Ribo-seq measures translation, which is not the same as protein level.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
Full-length cDNA from oligo-dT priming and template switching is amplified, made into circular templates and read by circular consensus, so each accurate long read is one complete transcript from poly(A) tail to 5′ end.
- Start from intact RNA (RIN ≥ 8); degradation creates false 5′ ends.
- Use the fewest PCR cycles that give enough cDNA, and consider adding back a long size-selected fraction.
- Keep only full-length non-chimeric reads before clustering into isoforms.
- Concatenation (array) approaches put several cDNAs in one read for more reads per run.