[Bioc-sig-seq] Illumina vs. ABI Solid
Wilhelm Brian
brian.wilhelm at umontreal.ca
Mon Jan 4 16:17:59 CET 2010
> On Dec 23, 2009, at 2:16 AM, Sulev Kõks wrote:
>
> > Hi,
> > It actually depends... I'm not so convinced on the superiority of
> Illumina. In fact, I know labs where Illuminas collecting dust.
> >
> > We use SOLID 3plus and are very satisfied. One very important issue for
> us - SOLID is the ONLY platform what allows to preserve strandedness. For
> any kind of RNA seq applications, this is actually cornerstone for the
> analysis of transcriptome. Without this information you just throw away
> substantial part of the data.
>
> This is not true. I know of at least two library preps for Illumina that
> preserve strandedness.
The maintenance of strandedness for RNA-seq is completely independent of machine type; as Kasper points out, there are several protocols for getting strand specific RNA-seq data using Illumina machines.
>
> > There is no price difference between platforms - but with one run in
> SOLID you get up to 60 GB of reads. Illumina doesn't reach this yet.
>
> There is a big difference between labs regarding how many reads they get.
> The market moves pretty fast. I am not too sure that SOLiD produces the
> most reads right now. And as far as I know they produce shorter reads,
> which is critical for some applications.
We have a SOLiD machine in place (3plus) and although it works well, we have never gotten close to getting the upper limit yields reported by either AB test or field labs. It works well when the sample is properly prepared and good quality. The development of read number and read length is fairly comparable between Illumina and SOLiD (at least close enough to not make a huge difference). AB has a kit for 50bp reads and will have a kit for 75bp soon (like Illumina currently has).
> > So, I do not think that Illumina is better than SOLID. There are labs
> who prefer Illumina, and labs who prefer SOLID. SOLID came out later and
> lost a lot of market share, but this is not the reason to think its worse
> that Illu.
>
> One major difference is that most tools right now operate directly on
> FASTQ files. This means that they do not operate in colorspace, and a big
> (possible) advantage of the SOLiD platform is lost. If you are in a lab
> with substantial bioinformatics resources this might not be a problem,
> because you can work around it. But if not, the selection of tools for
> the SOLiD platform is - in my impression - much smaller. And analysis is
> quickly becoming the the bottleneck for these types of data.
>
Our experience as well in analyzing SOLiD data is that although some software works on colourspace data as well as fastq files, having colourspace limits your options for analysis. It's not yet a problem in terms of being able to carry out analysis but some tools just can't be used.
It's worth keeping in mind also that the wet-lab sample prep time is shorter for the Illumina than for the SOLiD.
Brian
> Kasper
>
> > Cheers
> > Sulev
> >
> > On 23.12.2009, at 4:02, Steve Lianoglou wrote:
> >
> >> Hi,
> >>
> >> 2009/12/22 Droit Arnaud <Arnaud.Droit at ircm.qc.ca>:
> >>> Hi all,
> >>>
> >>> The institute where I currently work is in the process of buying a new
> HT sequencer. We are debating whether we should get an Illumina or a Solid
> sequencer. I was wondering if anyone of you could give me some advice in
> terms of analysis. As a bioinformatician using R, I am particularly
> interested in comparisons in terms of analysis, data manipulation/storage
> in R, etc. We have some experience analyzing Illumina data but no
> experience with Solid data.
> >>
> >> For whatever its worth, we had someone from The Broad recently give a
> >> talk about some work he's been doing with RNA-seq.
> >>
> >> Near the beginning of the talk he had a slide with the number of
> >> machines they had from each platform, it broke down to something
> >> roughly like so (if I remember correctly):
> >>
> >> Illumina: ~ 90
> >> SOLiD: < 10
> >> 454 : < 10
> >>
> >> And he mentioned that the solid and 454 machines are just basically
> >> collecting dust.
> >>
> >> I don't really have enough experience with either platform to draw and
> >> judgements myself, but the guy giving the talk seemed pretty convinced
> >> and thought it was relevant to share.
> >>
> >> -steve
> >>
> >> --
> >> Steve Lianoglou
> >> Graduate Student: Computational Systems Biology
> >> | Memorial Sloan-Kettering Cancer Center
> >> | Weill Medical College of Cornell University
> >> Contact Info: http://cbio.mskcc.org/~lianos/contact
> >>
> >> _______________________________________________
> >> Bioc-sig-sequencing mailing list
> >> Bioc-sig-sequencing at r-project.org
> >> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
> >
> > _______________________________________________
> > Bioc-sig-sequencing mailing list
> > Bioc-sig-sequencing at r-project.org
> > https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>
> _______________________________________________
> Bioc-sig-sequencing mailing list
> Bioc-sig-sequencing at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
More information about the Bioc-sig-sequencing
mailing list