Transcriptome data: detection of large genomic insert of unknown sequence

Question

[ashish]

I am doing transcriptome analysis of wheat translocation lines, means they have a "big chunk of DNA" inserted into their chromosomes. The plant from where this big chunk came is a wild plant with zero genome information available about it. This wild plant is actually a relative of wheat. I want suggestion about how do I know about the foreign "dna chunk" expression in my wheat translocation lines.. I think I should look into the unmapped reads file, assemble them and then use them as potential foreign transcripts. Please suggest me something better.

[Benn]

If your "DNA chunk" is from a relative, it might be that the genes on this DNA are very similar to those of the wheat plant. This means that with mapping to wheat genome, the transcripts from the chunk will be mapped to the genome as well. Thus using only unmapped reads will miss those.

Better would be to do trinity assembly with all reads, or try to sequence the DNA chunk first.

[cpad0112]

1. Amplify the fragment if possible
2. Send it for sequencing
3. Do a denovo assembly
4. Blast it
5. Once you find a hit, predict the possible transcripts/ORFs using known transcripts. If you do not find a hit, try to find ORFs from denovo assembly.

[ashish]

There is no genomic information available and the "dna chunk" is quite big which was incorporated into wheat using breeding. my last hope for sequencing it was using markers but that is also not available.

[ashish]

The "dna chunk" is from a relative. It is a wild relative of wheat. Common wheat does not have the genes present in that "dna chunk" so I don't think it will map but doing denovo assembly is a good idea. Thanks for the suggestion.

[WouterDeCoster]

how do I know about the foreign "dna chunk" expression in my wheat translocation lines..

What do you want?

[ashish]

I want to know how the foreign "dna chunk" is expressing in wheat. The foreign "dna chunk" comes from a grass which is related to wheat.

[WouterDeCoster]

So you want to know if it's expressed or not. Then I would search in the fastq files for an expected sequence from the "chunk". It may or may not get mapped properly.

[ashish]

But I don't know what sequence to search in the fastq file. This was done by breeding and the one I am using was screened using the phenotype. The person who did it determined using fluorescence microscopy that a "dna chunk" is inserted in chromosome 5 which is giving the phenotype. Sequence information is not available. What genes might be present in that "dna chunk" is also not known. The related species is a wild grass, no information is available about its sequence, markers or anything.

Answer 1

[Rohit]

If you have no genomic information, I think it might be hard to find the full piece of the added DNA. You can try to assemble the transcriptome denovo and map to the genome, in-order to check where you have inserts. But the problem here would be that its hard to be precise without long-read information, due to chimeric transcripts. Then map the unmapped reads to the transcripts in insert positions as a sanity check, and check for expression with a transcript-based approach.