Closed princy149 closed 1 year ago
hi @mortonjt,
could you suggest me about this? its highly helpful for analysis.
looking forward to hear from you,
If it is greater than 0, it is outperforming random. I wouldn't recommend any model fits that have a Q2 < 0
On Wed, Jul 20, 2022 at 9:20 AM UPASNA SRIVASTAVA @.***> wrote:
hi @[mortonjt]
I compiled songbird pipeline for my microbiome data based on different samples but due to low power of dataset, every time using different parameters like batch size, epoches, differential-prior values etc for different taxa level as Order and Genus , I got result with low values of Q2 as 0.014795 for one sample type and 0.056279 , 0.019888 for second sample type as shown in fitting model check. then is it fine if I move forward for my manuscript using these values?
please give me your suggestion on this,
— Reply to this email directly, view it on GitHub https://github.com/biocore/songbird/issues/165, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA75VXJRFFMSF72T7HWCH5DVU74IBANCNFSM54DUMWFQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>
hi @[mortonjt]
I compiled songbird pipeline for my microbiome data based on different samples but due to low power of dataset, every time using different parameters like batch size, epoches, differential-prior values etc for different taxa level as Order and Genus , I got result with low values of Q2 as 0.014795 for one sample type and 0.056279 , 0.019888 for second sample type as shown in fitting model check. then is it fine if I move forward for my manuscript using these values?
please give me your suggestion on this,