Hello,
I am teaching a Genomics Biotechnology class this semester at UNC Charlotte and have a question about the best alignment tool to use for a class project.
Students in the class are using Ion Torrent PGM to sequence the chloroplast genomes of several different cultivars of tomato. The students made libraries by amplifying chloroplast DNA (using long-range PCR), fragmenting the amplicons, ligating adapters, and then amplifying the libraries. In the next week, they'll sequence their libraries on the Ion Torrent PGM.
We have a tomato chloroplast reference sequence and would like to align the students' reads onto the reference. Recall that the chloroplast genome is circular.
Ion Torrent is prone to introducing lots of deletions and insertions next to runs of the same base. The data seem very similar to 454 data in that respect.
My question for you is:
- What alignment tool would you recommend we use?
We aligned data from previous years classes onto a reference chloroplast genome using bowtie2, but probably there are other tools that would work just as well, or even better.
Data from previous years classes are available in Integrated Genome Browser if you want to take a look. To see the data in IGB:
- Start IGB (go to bioviz.org to get a copy)
- Click the tomato shortcut image (on the left side of the Mona Lisa image, past rice and Arabidopsis)
- Select the chloroplast genome (in Current Sequence tab)
- Open and view data files in the folder named "Plastid Re-sequencing" (in Available Data section of the Data Access tab
Thank you!
Ann
BBMap is very easy to use and would be simple for students to use. BBMap package is a suite of tools that can do lots of cool things. It will work on PC/OS X/Linux being java.
Is it specifically designed for Ion Torrent reads? Any particular reason why it would be a better choice than say BWA or Bowtie2? Do you have to first QC the reads and then proceed with the alignment, or does it deal with QC on the go?
BBMap is a multi-purpose NGS aligner that can be used for various types of NGS data. It may (or may not) be a better choice than bwa or bowtie2. It is extremely easy to use (and fast). Brian Bushnell, developer of BBMap, is responsive to user queries/suggestions and provides helpful support in this and other forums.
I am not sure exactly what kind of QC you are referring to but it does have tools built in to do coverage calculation, trimming etc. Brian himself recommends FastQC if you are only looking to do QC.
In my tests, BBMap is more accurate when dealing with reads containing indels as compared to bwa/bowtie2, which should give it an edge when mapping Ion Torrent reads, as their error model includes indels.
What about IonTorrent's own tool, tmap? Is it no good?
tmap should be fine. I assume @Ann is trying to find if there is a specific alternative people recommend for Ion data.
Her point about students (who may be naive about NGS data/alignment) using the tool in class is important to remember. I don't know if the class users a central server/cluster to run the analysis or students are expected to install the software on their own machines, since that consideration may limit the choice of aligners that can be used.
In this case, this thread should be relevant (TLDR: they agree that tmap works fine) and also this benchmark. I think that tmap can be installed as standalone tool, just like bowtie2.