By Monica Heger
Expression Analysis, which provides sequencing services on the Illumina and Pacific Biosciences platforms, recently presented internally generated data using PacBio's new C2 chemistry that will be launched more broadly in the fourth quarter.
During a webinar, Pat Hurban, Expression Analysis's vice president of research and development, showed data from the sequencing of Escherichia coli using 2-kilobase inserts with the C1 chemistry and 6-kilobase inserts with both the C1 and C2 chemistries.
Despite working with the C2 chemistry for only about a week, "we are very, very happy with the results," Hurban said.
So far, the Expression Analysis team has seen average read lengths increase by 80 percent, to around 2,700 base pairs. Further optimization of the protocols, as well as a planned software upgrade from PacBio, should yield even more improvement, Hurban added.
Data from the C2 chemistry was first made available when PacBio released its results from sequencing the German E. coli outbreak strain (IS 7/12/2011). Using a combination of 9-kilobase inserts and short circular consensus sequencing, the company achieved average read lengths of 2,900 bases and a consensus accuracy of 99.998 percent
Expression Analysis, meantime, has so far only tested the chemistry with 6-kilobase inserts, but Hurban said the team plans to test insert sizes between 8 and 12 kilobases.
In the webinar, Hurban presented data on the sequencing of three E. coli strains on 14 SMRT cells with 2-kilobases inserts and two E. coli strains on 8 SMRT cells using 6-kilobase inserts, both using the C1 chemistry. Additionally, he showed data on the sequencing of two E. coli strains on 8 SMRT cells with 6-kilobase inserts using the C2 chemistry.
With the C2 chemistry, the Expression Analysis team achieved average read lengths of 2,715 base pairs, with a mean maximum read length of 13,091 base pairs. "Our personal best is just over 14,000 bases," Hurban said.
Compared to sequencing E. coli with the C1 chemistry using 6-kilobase inserts, the C2 chemistry upped average read lengths by 80 percent from 1,519 bases and doubled the average maximum read length from 6,512 bases.
The Expression Analysis team observed comparable single-pass accuracy between the C1 and C2 chemistries at around 85 percent, and consensus accuracy for both chemistries at 30-fold coverage at around 99.997 percent.
However, Hurban told In Sequence that Expression Analysis is still working on optimizing protocols for the C2 chemistry and is also waiting on software upgrades from PacBio that will be included with the commercial release of the C2 chemistry, which he said would further improve single-pass accuracy and throughput.
[ pagebreak ]
PacBio has said that the C2 chemistry will increase single-pass accuracy to 87 percent and double-pass circular consensus accuracy to 93 percent using a 1,350-base pair insert, and to 99 percent using a 500-base pair insert.
Hurban said that while Expression Analysis has not yet achieved these rates, "we anticipate that the C2 chemistry will have better single-pass accuracy once we obtain the software updates that are currently in development."
One interesting difference between the C2 and C1 chemistry that Hurban noted was that the ranges of accuracy rates and coverage were much narrower with the C2 chemistry. "It was an interesting and repeatable observation," he said, "but we don't fully understand why it is."
Hurban added that Expression Analysis has started offering the C2 chemistry to its customers for projects where it makes sense. "We want to make sure we have a proper fit between the specific project the customer is bringing to the table and the specific goals we have in terms of better understanding the C2 chemistry," he said.
Types of projects that would see immediate benefits from the C2 chemistry will be amplicon sequencing and small genome sequencing, where the longer read lengths will be extremely beneficial, Hurban said.
Going forward, he said the company would be looking to increase the insert sizes to between 8 kilobases and 12 kilobases. The increased insert sizes will be particularly beneficial for "small genome sequencing applications," and will produce "better assembly power."
Despite the potential of PacBio's platform to produce very long reads, which would enable better assemblies, adoption of the instrument since its launch in the second quarter of this year has been slower than expected. As a result, PacBio last week announced that it would have to lay off around 130 employees (IS 9/20/2011).
PacBio did not speculate as to why the instrument's uptake has been slower than expected, but following a second-quarter earnings call, analysts expressed concern about the instrument's high list price of $695,000, its accuracy, and long lead times to placements.
The RS has a higher single-pass error rate than other next-generation sequencing technologies, which has been a challenge for the firm despite the fact that circular consensus sequencing brings the error rate down so that it is comparable with other systems.
Nevertheless, users of the RS have reported positive results with the machine.
For example, the Ontario Institute of Cancer Research is using the system in a targeted sequencing pilot study of cancer patients with metastatic or recurrent disease (CSN 8/3/2011). John McPherson, who is leading the project, said that the data has been good, with long read lengths and a quick turnaround time.
Additionally, PacBio's CEO Hugh Martin said recently that users such as the Department of Energy's Joint Genome Institute and the Broad Institute have been running their machines "around the clock."
Have topics you'd like to see covered by In Sequence? Contact the editor at mheger [at] genomeweb [.] com.