Hi! I am asking because I had some luck using the SPAdes assembler to assemble the mitochondrial genome from subsamples of full genomic reads of parasitic nematodes, but it has not been successful for all samples. I am wondering if there are any circular contig assmblers available (for plasmids or plastid/mtDNA), other than the one available through geneious R7 (not freely available...). I have tried mitobim multiple times with many samples and have had very unsatisfactory results.
I have usually found that assembling the entire genome data works quite well, but with a slight change.
Try the MaSuRCA assembler - get to the superReads part..
1. install the assembler and run your data through the entire pipeline (http://www.genome.umd.edu/), OR
2. Shortcut: (http://ccb.jhu.edu/software/stringtie/index.shtml?t=manual) a wonferful script that utilizes this to the boot. Look at the "Assembling super-reads" section..
This should get you long reads that then can be subsampled using mitochondrial gene sequences from NCBI and assembled.
Maybe it would help to start your assembly with only mitochondrial reads? If so you could potentially use deconseq to filter out the nematode reads from your raw data (assuming you have a good reference genome with the mitochondrial contigs removed).
I say this because of my recent attempts to assemble a Canine Herpes virus genome. I used deconseq to remove all the canine genomic reads prior to assembly with MIRA, after the assembly I was left with some pretty nice mitochondrial contigs because my deconseq reference genome did not include mitochondrial sequences.
I had success assembling the full genomes and extracting the mtDNA. thank you!
@Artur what Matthew may have eluded to could be similar to the method used in http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3577829/ ..
Basically use the Denovo assembly tool and extract mitochondrial contigs..
Hello Guys! Normal denovo assembly algorithm of CLC genomics workbench also gives very good assembly as far as small mitochondrial genome is concerned (insects and animal mito genomes ~15Kb size) and then annotation can carried out using MITOS to identify the mito genes.
I found this thread as I was searching for suggestions for how to work with short Illumina reads to assemble mitogenomes. Caroline, it would be very helpful if you could share your solution here. It might not work for me but it is always useful to have additional input. I am currently attempting to assemble the mitogenomes of trout from Illumina reads and am using Velvet and an Oncorynchus mykiss reference genome without much luck.
Hi Emily,
Sorry for the long delay in response. I never found one successful method to this. I never got mitobim working properly, despite much effort. The best thing I did was use the SPAdes assembler, extract contigs of greater than 11,000 bp, BLAST them, and select the one that was the mtDNA. If I did not get contigs larger than 11,000 bp I took the contigs generated by SPAdes and assembled them again in SeqMan Pro (laser gene) to see if I could get longer contigs- then repeated the BLAST. Another, different method I used was to assemble the raw genomic reads in Geneious (R9), which can assemble circular contigs. In my case, the mtDNA is circular, so i would look for contigs of the appropriate size that were circular. If none existed I would also look for contigs of the correct size, even if they were linear, and BLAST them. To annotate mtDNA I used MITOS (http://mitos.bioinf.uni-leipzig.de/help.py). I did the assemblies in geneious de novo (no reference). I know this second method in geneious has worked for others working on mammalian DNA.
I hope this helps!
Caroline
It seems that a new assembly method for organelle genomes from whole genome data came out recently: NOVOPlasty (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5389512/).
- Badges
- Science method