Is it EBSeq-HMM appropriate for studying the stage specific expression profile in cancer?
38 views
Skip to first unread message
shaomin Wu
Mar 14, 2016, 7:59:51 AM3/14/16
to EBSeq users
Hi,
I am interest in stage specific expression profile genes in cancer. The vignette uses the time course data as example. For the tumor progression, I think that it's very similar with the ordinary time course data.
This is the condition level I proposed: "normal", "stage1", "stage2", "stage3", "stage4"
The sample number for each condition will be variable.
I like to ask if my experiment design fits the statistic model of EBSeq-HMM. Thank you
I am interest in stage specific expression profile genes in cancer. The vignette uses the time course data as example. For the tumor progression, I think that it's very similar with the ordinary time course data.
This is the condition level I proposed: "normal", "stage1", "stage2", "stage3", "stage4"
The sample number for each condition will be variable.
I like to ask if my experiment design fits the statistic model of EBSeq-HMM. Thank you
Ning Leng
Mar 14, 2016, 11:41:06 AM3/14/16
to shaomin Wu, EBSeq users
Hi Shaomin,
Yes EBSeqHMM can be applied to experiments with ordered conditions
other than time points. You'll be able to use it to identify genes
with different patterns ( monotone increasing, going up then down,
etc)
Best,
Ning
> --
> You received this message because you are subscribed to the Google Groups
> "EBSeq users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ebseq-users...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
--
Ning Leng, Ph.D
Computational Biologist / Biostatistician
Regenerative Biology Laboratory
Morgridge Institute for Research
Yes EBSeqHMM can be applied to experiments with ordered conditions
other than time points. You'll be able to use it to identify genes
with different patterns ( monotone increasing, going up then down,
etc)
Best,
Ning
> You received this message because you are subscribed to the Google Groups
> "EBSeq users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to ebseq-users...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
--
Ning Leng, Ph.D
Computational Biologist / Biostatistician
Regenerative Biology Laboratory
Morgridge Institute for Research
shaomin Wu
Mar 15, 2016, 10:02:09 AM3/15/16
to EBSeq users
Hi, Ning
I have use the this software to run my clinical data. But the result confuses me.
I plot the 50 genes with top confident "down-down-down" path, however, some of gene doesn't match to the path "down-down-down", see attached file.
Please give me some suggestions, thank you.
I have use the this software to run my clinical data. But the result confuses me.
I plot the 50 genes with top confident "down-down-down" path, however, some of gene doesn't match to the path "down-down-down", see attached file.
Please give me some suggestions, thank you.
Ning Leng
Mar 16, 2016, 11:10:24 AM3/16/16
to shaomin Wu, EBSeq users
Hi Shaomin,
Seems to me that there are large variabilities within condition (e.g.
outliers and perhaps subpopulations?) Then I suspect the model fitting
might not be good. What is the PP(down-down-down) range for these
genes? Although they are top down-down-down genes, it's possible that
they are not classified with high confidence.
You may try to remove samples that are quite different from the other
samples in the same condition? You may also try to set the FCV to a
larger number, e.g. 3. Then the dynamic genes will be expected to have
FC >=3. By doing so you may be able to get rid of genes with moderate
FC and high within condition variations.
Hope this helps,
Ning
Seems to me that there are large variabilities within condition (e.g.
outliers and perhaps subpopulations?) Then I suspect the model fitting
might not be good. What is the PP(down-down-down) range for these
genes? Although they are top down-down-down genes, it's possible that
they are not classified with high confidence.
You may try to remove samples that are quite different from the other
samples in the same condition? You may also try to set the FCV to a
larger number, e.g. 3. Then the dynamic genes will be expected to have
FC >=3. By doing so you may be able to get rid of genes with moderate
FC and high within condition variations.
Hope this helps,
Ning
Reply all
Reply to author
Forward
0 new messages