lhe17 / nebula

GNU General Public License v2.0
28 stars 6 forks source link

What is the normalization method used by NEBULA? #20

Open MANZHAOHUI opened 1 year ago

MANZHAOHUI commented 1 year ago

I am doing covariates selection analysis before the DE. Since this analysis should use the same normalization method with the actual DE analysis and NEBULA takes in the raw data, I am wondering what is the normalization method applied by NEBULA before the regression.
Thank you.

lhe17 commented 1 year ago

Hi ZHAOHUI,

"what is the normalization method applied by NEBULA before the regression."

NEBULA does not transform (or normalize) the expression value before the regression. Instead, the normalization is incorporated in the regression by adjusting the rate of the Poisson distribution through a scaling factor (e.g., cell library size). This is one of the key benefits of directly modeling the counts compared to data transformation.

However, NEBULA also offers Pearson residuals (https://github.com/lhe17/nebula#extracting-marginal-and-conditional-pearson-residuals) as normalized expression data for other analyses.

Best regards,

Liang

On 4/12/2023 8:44 PM, ZHAOHUI MAN wrote:

I am doing covariates selection analysis before the DE. Since this analysis should use the same normalization method with the actual DE analysis and NEBULA takes in the raw data, I am wondering what is the normalization method applied by NEBULA before the regression. Thank you.

— Reply to this email directly, view it on GitHub https://github.com/lhe17/nebula/issues/20, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGDISUSWF37CZOKFEBTSIGLXA3ZYLANCNFSM6AAAAAAW4B7LLE. You are receiving this because you are subscribed to this thread.Message ID: @.***>

MANZHAOHUI commented 1 year ago

Thank you very much. By the way, I am looking forward to your presenting NEBULA later this month in DUKE.

lhe17 commented 1 year ago

Hi ZHAOHUI,

That'll be great. We can discuss more in the meeting.

Best regards,

Liang

On 4/13/2023 9:38 PM, ZHAOHUI MAN wrote:

Thank you very much. By the way, I am looking forward to your presenting NEBULA later this month in DUKE.

— Reply to this email directly, view it on GitHub https://github.com/lhe17/nebula/issues/20#issuecomment-1507514682, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGDISUUKHFKIHZJHNTXD2DDXBBI2ZANCNFSM6AAAAAAW4B7LLE. You are receiving this because you commented.Message ID: @.***>