issues with degPatterns not setting Time in Normalize

[kesner]

First my versions: OSX latest R version:
[1] dplyr_1.1.1 tibble_3.2.1 plyr_1.8.8
[4] cluster_2.1.4 edgeR_3.40.2 limma_3.54.2
[7] DEGreport_1.34.0 pheatmap_1.0.12 ggplot2_3.4.2
[10] DESeq2_1.38.3 SummarizedExperiment_1.28.0 Biobase_2.58.0
[13] MatrixGenerics_1.10.0 matrixStats_0.63.0 GenomicRanges_1.50.2
[16] GenomeInfoDb_1.34.9 IRanges_2.32.0 S4Vectors_0.36.2
[19] BiocGenerics_0.44.0

I do not see Time being set in the structure returned from degPatterns.

metadata: Exp Time (integer 0,1,2) Treat (levels WT KO with WT being default)

clusters_rlog_padj<- degPatterns(cluster_rlog_padj, metadata = metadata, time="Time", col="Treat", minc=0)

In clusters_rlog_padj I see time set in raw but in normalized. To get degPlotCluster(clusters_rlog_padj$normalized, time="Time", color="Treat",min_genes=0) to work I first have to run clusters_rlog_padj$normalized$Time<-clusters_rlog_padj$raw$Time
I see this with View in Rstudio of the returned clusters_rlog_padj. Without doing this, normalized>Time has Type factor with values Factor with 0 levels. "". Even though Time is not set up as a factor but an integer taking the values 0 1 or 2.

As an aside I also see normalized>merge to be 'WT0' 'WT0' 'WT0' .... Where as in raw, merged is 'KO0' 'KO0' 'KO0' ...
Is this correct, should normalized had raw have the same merge order since time in Raw seems to be correct independent
of merged.

OTHER INFO:

metadata is
Exp Time Treat
s1 0 WT
s2 0 WT
s3 0 WT
s4 1 WT
s5 1 WT
s6 1 WT
s7 2 WT
s8 2 WT
s9 2 WT
s10 0 KO
s11 0 KO
s12 0 KO
s13 1 KO
s14 1 KO
s15 1 KO
s16 2 KO
s17 2 KO
s18 2 KO

Model using is : reduced: Treat + Time
full. Treat + Time + Treat:Time

Let me know what else you need?

[lpantano]

Hi,

Can you check Time column is a factor?

[lpantano]

As well, if you could save the object and send it to me, I could try to troubleshoot more.Thanks.

[kesner]

Hi, My Time is an integer in the raw version. In the normalized version it say it is a factor. I am doing time series analysis and want to use Time as part of the linear regression model as I thought your code is doing. Barry

…

On Thu, Apr 20, 2023 at 6:25 PM Lorena Pantano ***@***.***> wrote: Hi, Can you check Time column is a factor? — Reply to this email directly, view it on GitHub <#57 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEJYOYYSGEAE5OF56VOEEQ3XCGZUBANCNFSM6AAAAAAW2Q7HAI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[kesner]

Hi, Here is my code and data. Barry

…

On Thu, Apr 20, 2023 at 6:26 PM Lorena Pantano ***@***.***> wrote: As well, if you could save the object and send it to me, I could try to troubleshoot more.Thanks. — Reply to this email directly, view it on GitHub <#57 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEJYOYZIMMUG6G6JNZ7SHC3XCGZX7ANCNFSM6AAAAAAW2Q7HAI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[kesner]

Hi, Metadata,save needs to me stored as just metadata for the code to work. Barry

…

On Fri, May 5, 2023 at 4:00 PM Modo C ***@***.***> wrote: Hi, Here is my code and data. Barry On Thu, Apr 20, 2023 at 6:26 PM Lorena Pantano ***@***.***> wrote: > As well, if you could save the object and send it to me, I could try to > troubleshoot more.Thanks. > > — > Reply to this email directly, view it on GitHub > <#57 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AEJYOYZIMMUG6G6JNZ7SHC3XCGZX7ANCNFSM6AAAAAAW2Q7HAI> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> >

[kesner]

Another question. Do you know how to label the genes on the degPlotCluster plots? Barry

…

On Thu, Apr 20, 2023 at 6:25 PM Lorena Pantano ***@***.***> wrote: Hi, Can you check Time column is a factor? — Reply to this email directly, view it on GitHub <#57 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEJYOYYSGEAE5OF56VOEEQ3XCGZUBANCNFSM6AAAAAAW2Q7HAI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[lpantano]

Hi, I think you would need some ggplot code for that. Probably adding with geom_text or with ggrepel, if you don't have many it may work.

[kesner]

Thanks, Yes ilI did! Barry

…

On Thu, May 25, 2023 at 7:30 AM Lorena Pantano ***@***.***> wrote: Hi, I think you would need some ggplot code for that. Probably adding with geom_text or with ggrepel, if you don't have many it may work. — Reply to this email directly, view it on GitHub <#57 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEJYOY2AJY2Z2GZ3QKPPCQDXH47ENANCNFSM6AAAAAAW2Q7HAI> . You are receiving this because you authored the thread.Message ID: ***@***.***>