Announcement

Collapse
No announcement yet.

Sequel Sequencing Costs - Microbial Genomes

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Sequel Sequencing Costs - Microbial Genomes

    Howdy,

    Tried to post this once, and it disappeared.

    I have been tasked with assembling a per genome cost of Sequel sequencing (assuming we own the instrument already). We are primarily looking to close microbial genomes in the 2 - 8 Mbp range.

    I read 5-8 Gbp of output from a single Sequel run, and I'm also hearing we can multiplex 2 genomes per run pretty safely - seems to be we could do more, but maybe there's something I'm not seeing about the math or the actual "usable" percentage of the 5 Gbp estimated output... or it is a gross overestimate when used in the real world.

    I've heard $1000 / SMRT cell, but that doesn't include all the reagents necessary to make libraries, barcode, etc. Excluding the cost of DNA extraction (unless it is for some reason more expensive than Illumina/ONT), any help or insight that could be provided on running multiple microbial genomes on a single Sequel SMRT cell, primarily from a cost basis, would be greatly appreciated.

    Best,
    DocBio

  • #2
    Check with one of the PacBio authorized service providers.

    Comment


    • #3
      Will do, although not sure how forthcoming they'll be on their costs vs what they charge... some sequencing centers treat that as a bit of proprietary information.

      Just to be clear, I'm not looking for the charged cost, but what the cost of the consumables, library prep, SMRT cell, etc, typically run for people who are doing it in house, and what's reasonable in terms of number of say 5 Mbp genomes to fit on a run.

      Best,
      Docbio

      Comment


      • #4
        This most likely applies to PacBio as well, but every sequencing center may get different price breaks based on the "volume" of business they do with the company, so you may never be able to do an apples to apples comparison on internal real cost.

        If you don't have access to a Sequel "in house" then the best you may be able to do is get quotes from multiple providers. Devil is always in the details so compare the quotes carefully.

        Comment


        • #5
          If it helps: based on my expectation a more realistic expectation is 3-5 Gb per SMRT Cell (which, depending on no-one-knows-yet-why, occasionally can be as low as 0.5-1 Gb). I am not sure what you mean by the 'usable' percentage of this but these numbers reflect the subreads and not the 'scraps' as PB calls them.

          Regarding the number of microbes you can multiplex in the range of 2-8 Mb: easily more than 2 and preferably less than 6. The larger the number of species the more likely the chance of obtaining (very) non-uniformly distributed yield for each sample, which would lead to very fragmented assemblies for some of the samples and undesirably large coverage for some others, which itself can complicate assemblies.

          An idea is to ask for quotes for a SMRT Cell on an RS II and a Sequel from the same provider(s) to have an idea about how cheaper the internal cost might be in the case of Sequel. The difference should be reflective of the difference in the internal cost, which might come handy if you're more familiar with RS II than you are with Sequel.
          Last edited by Ali May; 06-22-2017, 04:31 AM.

          Comment


          • #6
            For comparison, the MinION consumable costs (in USD) are available on the ONT website. Here are some examples:

            Initial [capital] cost: $1000 (includes 2x flow cells, one sequencing kit for 6x runs, one wash kit for 12x runs)

            Flow cells: $900 (x1); $9480 (x12 @ $790); $16200 (24 @ $675); $24000 (48 @ $500)

            Standard Reagent kits: $510-$660 (6x runs)

            Rapid Barcoding kit: $672 (12 samples x 6 runs)

            Flow cell wash kit: $189 (allows flow cell to be re-used; wash buffer for 12 additional runs)

            Current yield is about 5-15Gb. There are a few labs that have successfully run >10Gb runs through careful sample prep, purification, and removal of short sequences (<500bp).

            Current accuracy is about 85-95%; consensus accuracy is about 97-99%. Accuracy will improve over time for existing sequencing runs/data via re-basecalling with the implementation of improved sequencing models.

            Comment


            • #7
              @gringer: In addition to the pure cost/yield for MinION can you comment on (if you have experience) chances of MinION data successfully able to "close" the microbial genomes? That appears to be a requirement in the original question.

              Based on data I have seen for PacBio the chances of that being achieved are pretty good. We are just getting out feet wet with MinION.

              Comment


              • #8
                Based on my experience using MinIon reads, especially those that are 2D, it is easier to close 2-10 Mb genomes when compared to Sequel. The error rate of MinIon is considerably lower. This said I'm not sure how relevant it is right now as it cannot be sold commercially.

                Originally posted by GenoMax View Post
                @gringer: In addition to the pure cost/yield for MinION can you comment on (if you have experience) chances of MinION data successfully able to "close" the microbial genomes? That appears to be a requirement in the original question.

                Based on data I have seen for PacBio the chances of that being achieved are pretty good. We are just getting out feet wet with MinION.

                Comment


                • #9
                  Nanopore isn't able to generate a full-length read per bacterial genome yet, but Nick Loman has had one run where he found 7 reads that covered the whole E. coli genome.

                  For single isolates, single-contig completion of bacteria and plasmids should be very easy with MinION long reads. A mixed sample will be a bit more difficult.

                  Comment


                  • #10
                    I do have access to a MinION, so I am familiar with it. Things are much much better with the R9.4 flow cells and we're still getting our protocols optimized for making the 1D libraries, interested in the 1D^2 library process. I am pleased with how the technology is progressing, but thus far have still had difficulty closing microbial genomes with just MinION data, or even hybrid assemblies with MinION and Illumina. Some of this is likely on us in terms of not getting the "best" results out of MinION and still learning the Canu and other assemblers.

                    I will say that my impression of PacBio as being a "more mature" or "more reliable" platform compared to ONT has been somewhat reshaped as I start talking to PacBio users.

                    Since another department is purchasing the Sequel, it may be that I find myself having access to both.

                    I appreciate all the input.

                    Comment


                    • #11
                      We released a protocol and barcoding kits this past spring for multiplexing up to 30 Mb of microbial genomes on the Sequel System – depending on the genome size that’s between 3-16 microbes per Sequel SMRT Cell, but on average probably 6. With 6 pooled microbes, the per microbe cost is $185 in reagents. That includes library prep with barcodes from the 8A / 8B barcode kits and sequencing with v2.1 chemistry including all consumables. With 10 microbes, the price drops to $120 / microbe, and with 16-plex it drops to $95 / microbe. You can expect QV40 assemblies using HGAP4. See the poster from the recent ASM-NGS mtg for more information.

                      Comment

                      Working...
                      X