[Jeremy]

How can you reconstruct the genome using transcriptome data?

[francicco]

How can you reconstruct the genome using transcriptome data?

Well, definitely not. But mtDNA is quite smaller than the nuclear genome; it's a single molecule while nuDNA has chromosomes; mtDNA genes do not splice or have different isoforms. I'd say is a totally different approach.

Still I do not see contraindications, what do you think is the misleading effect?

F

[francicco]

...to assemble all these condings (one for each read block) and obtain a sort of consensus.

Do you have any advise how to do it? I was trying to use CAGOB (wgs-assembler), but since contigs have "x" (gaps) apparently it does not work!

Anyway, any help about this issue?

F

[Jeremy]

If you are talking about a mammalian mitochondrial genome maybe. Plant mitochondrial genomes are very very different.

[francicco]

If you are talking about a mammalian mitochondrial genome maybe. Plant mitochondrial genomes are very very different.

I'm trying to do it on a Drosophila mtDNA.

[bioman1]

My thoughts to get Drosophila mtDNA
1. De novo assemble all your RNA seq transcripts using velvet/oases, soap de novo, trinity or any other your favourite transcriptome assembler.
2. Blast the assembled transcripts against insect mitochondrial genes and extract the reads.
3. Another way, to get search for Cytochrome c oxidase I (COX1) sequences between your RNA seq reads and closest reference genome. Then do reference based genome assembly.

[Jeremy]

The topic of this thread was plant mitochondrial genome assembly. In any case, have you subsetted your data to get only the mitochondrial sequence? Based on the number of reads you are talking about it sounds like maybe you haven't? You can either subset only the mitochondrial reads for a genome assembler or run the whole thing through a transcriptome assembler then subset the mitochondrial contig(s).

[francicco]

Thank you very much guys.
F

[bioman1]

I have tried MITObim tool to extract mtgenome, I got lot of gaps in mtgenome assembly. So I tried to do de novo assembly of my WGS data and extract mtcontigs through blast. Below are the steps

1.I have de novo assembled WGS illumina reads (2x101 bp) using CLC workbench. Identified mitochondria reference genome by blasting (blast N) my de novo assembled genome against NCBI plant mitochondrial genomes (http://www.ncbi.nlm.nih.gov/genomes/...&opt=organelle) and selected reference genome which has top hits in blastn (e.g papayamt genome).

2. Then extracted contigs from my denovo assembly which has more than 80% identity against papayamt genome. I have around 163 contigs which range from 200 bp to 2 Kb.

How to further process the extracted contigs?. Do I have to keep only longer length contigs (larger than 1Kbp)?. How to assemble single circular genome as mentioned in published papers?

[lu_ma]

hi, Bioman
Have you got de novo assembled mitochondria in your plant from hi-seq 2000data?
I am doing the similar job and found it is not easy because of the huge size of mitochodria in plant.
Would you share some experience if you have got anything.
Thanks a lot

Hi all,

I am trying to assemble plant mitochondria genome. The method I follow is to extract mitochondria reads from genomic reads (sequenced WGS approach using hiseq 2000, illumina paired-end reads)

1. I have downloaded all mitochondrial genomes of plants and indexed as reference genome using BWA
2. The raw paried-end reads were filtered (adapter & low quality reads filtered) which passed fastqc tool test. The fastqc passed filtered reads were interleaved using using perl script and used as single-end sequence. These single-end sequence were mapped to mitochondiral reference genome using BWA
3. Then mapped reads are extracted using samtools -F 4 option and got output in bam format
4. Using picard, bam format converted to fastq format
5.Before doing denovo assembly, I checked with fastqc, it failed in following
(i)FAIL-Per sequence GC content
(ii)FAIL-Sequence Duplication Levels
(iii)FAIL-Overrepresented sequences
(iv)FAIL-Kmer Content

My questions
(i) what I can I improve the reads before denovo assembly of mitochondrial reads?
(ii) Which better tool to assembly mitochondrial genome velvet or soapdenovo?. How much kmer size can be used?