Why would be de novo gene assembly from RNA-seq data useful?

Question

[Yugalee]

I've read some papers where researchers try to assembly genes by just using RNA-seq data. Why would this be useful? Why not just map the RNA-seq to the genome and then quantify and detect genes?

[Nicolas Rosewick]

The first application is when you don't have a reference genome to align on.

[jeremias.br]

As mentioned by Nicolas there are many situations where a good reference is not available. This is especially true in non model organisms. In many cases it might be possible to map to a closely related reference but you will miss novel genes. So de novo assembly is also usefull to detect those novel genes.

[Kevin Blighe]

• Where no reference transcriptome or genome is available (Nicolas)
• To search for novel genes/transcripts, including non-coding RNAs (jeremias)
• New splice isoforms of already identified genes (or the newly identified transcripts)
• To search for evidence of fusion-genes