Is there any literature or so that you can point to, which could give me the idea of comparing the cloud based genomic platforms (such as DNAnexus, BaseSpace, Seven Bridges etc.) in terms of functionalities/ costing/ storage/ integration of public bioinformatic databases as well as some analytic tools?
ADD COMMENT
• link
updated 20 months ago by
Ram
43k
•
written 8.5 years ago by
asmi
▴
40
2
Entering edit mode
I am not familiar with all companies on the list you have referred to above but I assume some use proprietary software where as others use open-source.
I see that Google Compute, Amazon AWS (and Microsoft azure) are not included in that list. You can rent commodity servers from them and run your own pipelines/workflows so they can also be cloud platforms for genomic analyses. In fact, google does have a dedicated offering for genomics: https://cloud.google.com/genomics/
What is it that you want to accomplish using the cloud?
I am looking into solving variant analysis and classification for private genomic data. I was wondering if such workflow / pipeline for analysis is already been made. The ones I mentioned - DNAnexus, BaseSpase - they all have their own genomic analysis workflows developed on top of AWS.
ADD REPLY
• link
updated 4.4 years ago by
Ram
43k
•
written 8.5 years ago by
asmi
▴
40
1
Entering edit mode
If you want to have flexibility (and it sound like you have the necessary IT expertise) rolling your own pipeline using standard software (e.g. aligner of choice, GATK, picard, freebayes etc) will give you lot of leeway. If you would rather not delve into the hassle of maintaining your own pipelines then using one of the commercial packages is logical. Give the ones you can, a try (before you buy), to see what would best fit your needs/budget.
Hi Asmi, please update your post to include what you are specifically wanting to do or find out. For example, do you want to run your software/analysis yourself or have someone do an analysis for you? That will help you get a better answer and save people time. If you just want to compare prices/services, you can obviously do that in a few minutes of web searching following the link you provided (I'm sure no one will want to do that for you). I don't think you will find a publication if that is what you want because this information changes often.
Hello SES, I am really new to NGS. And so I am trying to explore both the cases - writing my own pipeline (if need be, when there is nothing that is already been done) or have it done by, for example: DNAnexus on top of AWS (majority of the proprietary ready made pipelines are built over AWS it seems). That is the reason I turned here to ask my question, to get some advice from the community. As I said, any help is appreciated! Of course, I understand no one would google and reply with a comparison table for me. Thanks anyway, for taking time to reply. Much appreciated.
ADD REPLY
• link
updated 4.4 years ago by
Ram
43k
•
written 8.5 years ago by
asmi
▴
40
The various cloud genomics providers, offer different things, with some overlap. For instance, DNANexus build on AWS platform, which is further built on by SaaS providers such as Tute Genomics. Most of the underlying code is open-source, but the integration of the pieces of software and implementation within a given infrastructure is where the complexity is.
If you're looking for a turn-key cloud-based bioinformatics solution for research genomics, there are a variety of providers. However, in the clinical space it will depend on your local regulatory and organisational requirements as there are few solutions directly marketed for clinical use.
As was mentioned, you really need to define what your needs are, and then demo a bunch of different solutions. All of the solutions will be evolving and I guess the advantage of going with the cloud is that switching from one provider to another may not be so difficult, assuming you don't require too much customisation of integration.
I am not familiar with all companies on the list you have referred to above but I assume some use proprietary software where as others use open-source.
I see that Google Compute, Amazon AWS (and Microsoft azure) are not included in that list. You can rent commodity servers from them and run your own pipelines/workflows so they can also be cloud platforms for genomic analyses. In fact, google does have a dedicated offering for genomics: https://cloud.google.com/genomics/
What is it that you want to accomplish using the cloud?
I am looking into solving variant analysis and classification for private genomic data. I was wondering if such workflow / pipeline for analysis is already been made. The ones I mentioned - DNAnexus, BaseSpase - they all have their own genomic analysis workflows developed on top of AWS.
If you want to have flexibility (and it sound like you have the necessary IT expertise) rolling your own pipeline using standard software (e.g. aligner of choice, GATK, picard, freebayes etc) will give you lot of leeway. If you would rather not delve into the hassle of maintaining your own pipelines then using one of the commercial packages is logical. Give the ones you can, a try (before you buy), to see what would best fit your needs/budget.
Hi Asmi, please update your post to include what you are specifically wanting to do or find out. For example, do you want to run your software/analysis yourself or have someone do an analysis for you? That will help you get a better answer and save people time. If you just want to compare prices/services, you can obviously do that in a few minutes of web searching following the link you provided (I'm sure no one will want to do that for you). I don't think you will find a publication if that is what you want because this information changes often.
Hello SES, I am really new to NGS. And so I am trying to explore both the cases - writing my own pipeline (if need be, when there is nothing that is already been done) or have it done by, for example: DNAnexus on top of AWS (majority of the proprietary ready made pipelines are built over AWS it seems). That is the reason I turned here to ask my question, to get some advice from the community. As I said, any help is appreciated! Of course, I understand no one would google and reply with a comparison table for me. Thanks anyway, for taking time to reply. Much appreciated.