comments on "Partek Flow" ?

[Luke97]

Hi, I used Partek Flow and this product was not that great and very expensive for the current features available on the product. It seems this product is not ready for the market yet, to many problems. I used CLC Bio with excellent results and they now give full local support across Europe.

[Weipingchen]

Hi,
This is a good question. I have asked the similar question when I was introduced to the Partek Flow. I think that Partek Flow is a just new web-based product that makes use of multiple CPU's which is useful for compute-intensive tasks often encountered in NGS data. It provides various annotations for various purposes or from databases. Partek's Genomics Suite has a bigger suite of downstream analysis statistical tools, but it does not have the GSA. So I think this may be case dependent, it depends on what type of analysis is most important for you. We use Partek Genomics Suite a lot for the downstream analysis and visualization for both Microarray and NGS data in our core lab daily, such as clustering, pathways and gene ontologies, etc. I just started to work on raw sequencing data from different platforms by Partek Flow for re-annotation purpose and etc. I think system biologist would love to have it for looking the sequence data again from ways with new questions which is Genomics Suite may not have in term of with new annotation from a new database and etc. You may then use the Genomics Suite to analysis the results again from the Flow. Best.

[secda1]

Hi,

Thanks for all the infos. One more question: would you then recommend using Partek Flow with the Gene specific analysis tool rather than the "normal" Genomic suite ?

[Weipingchen]

About Parteck Flow's GSA

Hi there, Greetings. Our core facility is using Partek Genomics Suite, and I’ve tried Partek Flow's GSA (Gene Specific Analysis) for differential gene expression and I think this may be by far the best out there for my limited knowledge for followings: First of all, it tests for log normal, Poisson, negative binomial, zero-inflated Poisson, and zero-inflated negative binomial for each gene which is especially helpful for analyzing clinical related data. Second, it checks multiple factors for each gene (e.g. tumor status, gender, etc.) to see which factors are influencing that particular gene. It thus has at lease two obvious benefits: A) it increased power due to better model fit, and B) it may help us to learn more about the biology, such as, which factors are influencing your gene expression in your experimental designs. All of the other algorithms you mentioned assume that there is only one single distribution apply to all genes and that all genes are influenced by the same factor, which may not be true in a lot of cases. In addition, Partek Flow's GSA can handle any number of factors. Many of the public domain algorithms seems to handle only one factor. I hope above may help parts of your questions. Best, Weiping Chen.

[secda1]

Hi all,

I was just wondering if anybody had tried the new software from Partek: "Partek flow" for differential gene expression?
And the pros/cons it had versus the usual software such as cufflinks, EdgeR, DeSeq ... ?

Cheers