JGI Researchers on The Economics of New PacBio Instrument
A month back, Pacbio announced the new instrument and we wrote -
I have been trying to do some back-of-the-envelope calculation of how much it would cost to assemble human-sized genomes with the new instrument and get less than $1000. Is that right?
A more detailed answer comes from an informative post from researchers at JGI, who are involved in large-scale genome sequencing.
Main points of the linked article -
1. JGI does a lot of sequencing, and cost matters to them.
2. They are among the lucky ones receiving one Pacbio instrument (Sequel) this year.
3. Here was the situation prior to new Pacbio and before considering GC-bias issue -
Producing the long mate pair library and sequencing on Illumina costs us about the same as sequencing on the PacBio RSII, while a PacBio 10kbp library produces a better assembly overall with the same current throughput.
3. After the new Pacbio machine -
Given that we already use ~5 SMRT cells for a eukaryotic genome, switching to the PacBio Sequel will likely mean that this product type will become even more cost effective. This is calculated using the announced numbers: ~7X the GBp per SMRT cell at 2-3X the SMRT cell cost would mean approximately ~2X cost saving.
4. Moreover, some genomes cannot be sequenced using Illumina at all -
An added advantage…is that PacBio sequencing displays no GC-bias, so this is an additional advantage for organisms with very high or low GC genomes. An example is the JGI recently completed sequencing and assembly of a Piromyces fungal genome with <20% GC which, for the last decade, has resisted all previous attempts at sequencing and assembly.
5. Mick Watson needs to learn bioinformatics 101.
An additional possibility that would streamline laboratory preparation, and therefore cost, is to skip barcoding entirely.
…………….
I had a short twitter conversation and this approach with Mick Watson who outlined some possible problems.
6. Cost of Illumina + Xanax is still less than the cost of Pacbio sequencing (summary of the last section).
You can read the full post here.
