negative cell type proportions?

[amr15]

Hi,
I had a quick question -- I was wondering if you could help me understand what negative cell type proportions mean in the results?

Thanks,
Anu

[mchikina]

The results are unit-less, they are only meant to correlate with cell-type proportions. That is why they are called "surrogate". For the purpose of most downstream statistical test and manipulations (t-test, interaction tests) having the correct unit wouldn't change the results.

…

On Tue, Mar 10, 2020 at 11:30 AM amraman ***@***.***> wrote: Hi, I had a quick question -- I was wondering if you could help me understand what negative cell type proportions mean in the results? Thanks, Anu — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACMSGDCREADUMZSEXEH6FMLRGZMLBANCNFSM4LFBUPEA> .

-- Maria Chikina University of Pittsburgh Computational and Systems Biology Phone: (412) 648-3338

[amr15]

I see—so if I don’t have gold standard data to correlate these results with , then would it be accurate to say that negative values indicate cell types that are depleted in the bulk sample? Thanks! Anu

On Tue, Mar 10, 2020 at 11:34 AM mchikina ***@***.***> wrote: The results are unit-less, they are only meant to correlate with cell-type proportions. That is why they are called "surrogate". For the purpose of most downstream statistical test and manipulations (t-test, interaction tests) having the correct unit wouldn't change the results. On Tue, Mar 10, 2020 at 11:30 AM amraman ***@***.***> wrote: > Hi, > I had a quick question -- I was wondering if you could help me understand > what negative cell type proportions mean in the results? > > Thanks, > Anu > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > < #1 >, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/ACMSGDCREADUMZSEXEH6FMLRGZMLBANCNFSM4LFBUPEA > > . > -- Maria Chikina University of Pittsburgh Computational and Systems Biology Phone: (412) 648-3338 — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA7QDY4LLPJCDNYLD2KTRDTRGZMZLANCNFSM4LFBUPEA> .

-- Anugraha M Raman P: (774) 286-9083 | E: amraman@post.harvard.edu

[mchikina]

Yes, exactly. Depleted relative to the mean.

…

On Tue, Mar 10, 2020 at 12:42 PM amraman ***@***.***> wrote: I see—so if I don’t have gold standard data to correlate these results with , then would it be accurate to say that negative values indicate cell types that are depleted in the bulk sample? Thanks! Anu On Tue, Mar 10, 2020 at 11:34 AM mchikina ***@***.***> wrote: > The results are unit-less, they are only meant to correlate with cell-type > proportions. That is why they are called "surrogate". For the purpose of > most downstream statistical test and manipulations (t-test, interaction > tests) having the correct unit wouldn't change the results. > > On Tue, Mar 10, 2020 at 11:30 AM amraman ***@***.***> wrote: > > > Hi, > > I had a quick question -- I was wondering if you could help me understand > > what negative cell type proportions mean in the results? > > > > Thanks, > > Anu > > > > — > > You are receiving this because you are subscribed to this thread. > > Reply to this email directly, view it on GitHub > > < > #1 > >, > > or unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/ACMSGDCREADUMZSEXEH6FMLRGZMLBANCNFSM4LFBUPEA > > > > . > > > > > -- > Maria Chikina > University of Pittsburgh > Computational and Systems Biology > Phone: (412) 648-3338 > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > < #1 >, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AA7QDY4LLPJCDNYLD2KTRDTRGZMZLANCNFSM4LFBUPEA > > . > -- Anugraha M Raman P: (774) 286-9083 | E: ***@***.*** — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACMSGDDBI3K76TBGY2ZEIFTRGZUYJANCNFSM4LFBUPEA> .

-- Maria Chikina University of Pittsburgh Computational and Systems Biology Phone: (412) 648-3338