z*8
2 楼
用过了，感觉不错，就是不知道那个negative binomial model是个什么咚咚。反正我
也不发bioinfo方面的杂志，应该不需要知道的太多。嘿嘿
也不发bioinfo方面的杂志，应该不需要知道的太多。嘿嘿
x*m
3 楼
就是nb啊，不做bioinfo，知道他nb就行了。：）
【在 z*********8 的大作中提到】
: 用过了，感觉不错，就是不知道那个negative binomial model是个什么咚咚。反正我
: 也不发bioinfo方面的杂志，应该不需要知道的太多。嘿嘿
【在 z*********8 的大作中提到】
: 用过了，感觉不错，就是不知道那个negative binomial model是个什么咚咚。反正我
: 也不发bioinfo方面的杂志，应该不需要知道的太多。嘿嘿
N*n
4 楼
是R package? 除了能找differential gene外，还要什么用处？
l*1
5 楼
pls refer
>
One issue with RNA-seq data, however, is that the variance of this
probability among different individuals of a group is substantially higher
than the mean, with respect to many genes (Anders, Huber 2010). A Poisson
distribution assumes an equal mean and variance, and is therefore not a good
fit. This issue, known as “overdispersion,” has inspired statistical
software authors to adopt other models, particularly the negative binomial (
NB) distribution, which is characterized by an additional dispersion
parameter. Several popular differential expression packages, such as edgeR (
Robinson, McCarthy, Smyth 2010) and DESeq (Anders, Huber 2010) are based on
the NB distribution, but they differ extensively in how the dispersion
parameter is estimated, how normalization is performed, or how the
hypothesis test is carried out. For a nice tutorial on how DESeq works in
these respects, and its actual usage, see: cgrlucb.wikispaces.com/Spring+
2012+DESeq+Tutorial.
http://rnaseq.uoregon.edu/analysis.html
original hint was from former post here:
同主题阅读：简单介绍 Bioinformatics Tools for NGS 分析
[版面:生物学][首篇作者：jcp] , 2011年09月10日
http://www.mitbbs.com/article_t/Biology/31569333.html
【在 N******n 的大作中提到】
: 是R package? 除了能找differential gene外，还要什么用处？
>
One issue with RNA-seq data, however, is that the variance of this
probability among different individuals of a group is substantially higher
than the mean, with respect to many genes (Anders, Huber 2010). A Poisson
distribution assumes an equal mean and variance, and is therefore not a good
fit. This issue, known as “overdispersion,” has inspired statistical
software authors to adopt other models, particularly the negative binomial (
NB) distribution, which is characterized by an additional dispersion
parameter. Several popular differential expression packages, such as edgeR (
Robinson, McCarthy, Smyth 2010) and DESeq (Anders, Huber 2010) are based on
the NB distribution, but they differ extensively in how the dispersion
parameter is estimated, how normalization is performed, or how the
hypothesis test is carried out. For a nice tutorial on how DESeq works in
these respects, and its actual usage, see: cgrlucb.wikispaces.com/Spring+
2012+DESeq+Tutorial.
http://rnaseq.uoregon.edu/analysis.html
original hint was from former post here:
同主题阅读：简单介绍 Bioinformatics Tools for NGS 分析
[版面:生物学][首篇作者：jcp] , 2011年09月10日
http://www.mitbbs.com/article_t/Biology/31569333.html
【在 N******n 的大作中提到】
: 是R package? 除了能找differential gene外，还要什么用处？
l*1
6 楼
so what? if your found can solve solid wet or hard bio can’t solve non-
linear trend between
count level and variance by Deseq or other softs ?
pls refer,
01-16-2013, 12:40 AM #1
JesperGrud
Junior Member
Location: Odense
Join Date: Aug 2012
Posts: 5
DESeq and independent filtering
Hi everyone
I know this topic has been up a few times, but yet there is a question. So
the basic idea about filtering is that it is done unsupervised to remove
genes that are too lowly expressed to become significant. This will in turn
reduce the number of tests made and therefore improve the multiple testing
correction of Benjamin-Hochberg. This is basically what Bourgon 2010 finds.
In the DESeq vignette, it is suggested to filter on the sum of reads for
each gene and remove the ones in the 40% bottom quantile. It is here the
question comes. Those 40% seem rather arbritary to me and must depend on the
data set.
So the question is if it statistically sound to just iterate through say 20-
60% cut-off and determine, which yields the best statistics and just use
that?
----
06-06-2013, 10:44 PM #70
sdriscoll
Senior Member
Location: La Jolla, CA, USA
Join Date: Sep 2009
Posts: 332
I do wonder why low-expressed genes tend to get the short end of the stick
in filtering. They aren't irrelevant...they are obviously expressed for some
reason unless we want to argue that we've got extra genes expressed doing
nothing at all. One explanation I tend to use is that when their count
values are low and additionally their coverage is very low across all
samples it's hard to say whether what we're seeing is noise or real evidence
that the gene is present but expressed very low. We can look at it as a
technical limitation of the sequencing run - we just didn't get enough reads
to test those genes.
Back to the filtering I do wonder one thing. Low count features tend to have
very high coefficients of variation but very low variance values. Highly
expressed genes tend to have very small coefficients of variation but very
high variance values. Is this maybe why Simon says they use the means
instead of the variances in their adaptation of what's outlined in the paper
? I fail to see how a linear cutoff could be applied when there's such a
clear non-linear trend between count level and variance.
http://seqanswers.com/forums/showthread.php?t=26560
【在 x******m 的大作中提到】
: 就是nb啊，不做bioinfo，知道他nb就行了。：）
linear trend between
count level and variance by Deseq or other softs ?
pls refer,
01-16-2013, 12:40 AM #1
JesperGrud
Junior Member
Location: Odense
Join Date: Aug 2012
Posts: 5
DESeq and independent filtering
Hi everyone
I know this topic has been up a few times, but yet there is a question. So
the basic idea about filtering is that it is done unsupervised to remove
genes that are too lowly expressed to become significant. This will in turn
reduce the number of tests made and therefore improve the multiple testing
correction of Benjamin-Hochberg. This is basically what Bourgon 2010 finds.
In the DESeq vignette, it is suggested to filter on the sum of reads for
each gene and remove the ones in the 40% bottom quantile. It is here the
question comes. Those 40% seem rather arbritary to me and must depend on the
data set.
So the question is if it statistically sound to just iterate through say 20-
60% cut-off and determine, which yields the best statistics and just use
that?
----
06-06-2013, 10:44 PM #70
sdriscoll
Senior Member
Location: La Jolla, CA, USA
Join Date: Sep 2009
Posts: 332
I do wonder why low-expressed genes tend to get the short end of the stick
in filtering. They aren't irrelevant...they are obviously expressed for some
reason unless we want to argue that we've got extra genes expressed doing
nothing at all. One explanation I tend to use is that when their count
values are low and additionally their coverage is very low across all
samples it's hard to say whether what we're seeing is noise or real evidence
that the gene is present but expressed very low. We can look at it as a
technical limitation of the sequencing run - we just didn't get enough reads
to test those genes.
Back to the filtering I do wonder one thing. Low count features tend to have
very high coefficients of variation but very low variance values. Highly
expressed genes tend to have very small coefficients of variation but very
high variance values. Is this maybe why Simon says they use the means
instead of the variances in their adaptation of what's outlined in the paper
? I fail to see how a linear cutoff could be applied when there's such a
clear non-linear trend between count level and variance.
http://seqanswers.com/forums/showthread.php?t=26560
【在 x******m 的大作中提到】
: 就是nb啊，不做bioinfo，知道他nb就行了。：）
相关阅读
请教一个关于isolation of chromatin from nuclei的问题能否同时申请两个以上职位？paper help, thanks!Postdoc Interview 之后有几篇中文文章，要不要写在CV里？哪里能找到不要求申请者有绿卡的小grant？有人用过clontech的ZsGreen1 GFP吗？请问大家医院里面有没有四五十岁了还整天做手术的阿。跑蛋白胶的问题paper help, thanks!问俩技术问题--laser capture microscopy & microarray怎么找两个物种间的同源基因探讨一下，四五十岁以后大家应该做什么工作才体面啊？饶毅对海归归海的博文 (转载)我不是投诉，只想问斑竹对版规的看法投了一篇文章，4周时问，说没收到第二个review，现在5.5周再问的话，做科学的原动力在哪里饶毅施一共反对的大科学来了！Life Sciences Salary Survey, 2009别人写信老叫我Dr怎么办