(with Adina Howe, James Tiedje, Titus Brown)
I have been working on the assembly of big shotgun metagenomic data from ARMO (Amazon Rain Forest Microbial Observatory) project. The biggest challenge is the huge data size, 2TB in fastq and more than 6 billions reads after read trimming. One lucky thing …
read moreThere are comments.
I gave a talk last Wednesday at U. Michigan in the DCMB program where I included a slide estimating how much DNA sequencing (in base pairs) was needed for good de novo assembly of sequences from various biological environments or problems. The slide was there to motivate the challenges of …
read moreThere are comments.
These talk notes are for my talk at the 2012 Argonne Soil Metagenomics Workshop.
The slides are available for viewing and download here, on Slideshare.
I'm going to be talking about our assembly pipeline for soil metagenomes.
Much of this work was done by …
There are comments.
One of the biggest problems with basic sequence analysis -- some would say the biggest problem -- is the error rate. If our sequencing reads were error-free, both assembly and mapping would be much, much easier. Alas, Illumina reads have a 0.1-1% error rate per base, and PacBio has an error …
read moreThere are comments.
I'm at the MBL STAMPS course, "Strategies and Techniques for Analyzing Microbial Population Structure," and one of the things I needed to address in my morning talk was the role that the k parameter plays in de Bruijn graph assemblers.
In most de Bruijn assemblers that I have used -- Velvet …
read moreThere are comments.