how to set library size in DESeq?
1
0
Entering edit mode
wang peter ★ 2.0k
@wang-peter-4647
Last seen 10.2 years ago
how to set library size in DESeq? i know how to set library size in edgeR d <- DGEList(counts = d, lib.size = c(9893630,11055814,11207084,9663487,11455088,8140053), group = group)#??DGEList?? but donot know how to set library size in DESeq it seems i can set size factor sizeFactors(object) <- value or pData(cds)$sizeFactor <- value but how to set library size? -- shan gao Room 231(Dr.Fei lab) Boyce Thompson Institute Cornell University Tower Road, Ithaca, NY 14853-1801 Office phone: 1-607-254-1267(day) Official email:sg839 at cornell.edu Facebook:http://www.facebook.com/profile.php?id=100001986532253
• 3.6k views
ADD COMMENT
0
Entering edit mode
Simon Anders ★ 3.8k
@simon-anders-3855
Last seen 4.3 years ago
Zentrum für Molekularbiologie, Universi…
On 02/15/2012 04:24 AM, wang peter wrote: > how to set library size in DESeq? > i know how to set library size in edgeR > d<- DGEList(counts = d, lib.size = > c(9893630,11055814,11207084,9663487,11455088,8140053), group = > group)#??DGEList?? > but donot know how to set library size in DESeq > it seems i can set size factor > > sizeFactors(object)<- value or pData(cds)$sizeFactor<- value > but how to set library size? > The size factor is all there is to set. For the analysis, the algorithm only needs to know the size of a library _relative_ to the other libraries. The total library size is irrelevant. In edgeR, the information is split over two field, one with the library sizes, and one with normalization factors. If you multiply the library sizes with the normalization factors (and divide by some constant to get values close to one), you get edgeR's equivalent to DESeq's size factors, and (to my knowledge) this is all that edgeR uses internally. Why would you want to set the library size? Simon
ADD COMMENT
0
Entering edit mode
dear Dr. Simon thank you very much for your clear explanation. i want to set the library size as the mapped read number. so i can define any one sample as one, and then others relative ratios. is it right to set the size factor? shan gao
ADD REPLY
0
Entering edit mode
Hi On 02/15/2012 02:57 PM, wang peter wrote: > thank you very much for your clear explanation. > i want to set the library size as the mapped read number. > so i can define any one sample as one, and then > others relative ratios. is it right to set the size factor? I still do not know _why_ you would want to set the size factors. In a standard analysis, this should not be done. Rather you should use the estimators provided by the software (function 'estimateSizeFactors' in DESeq, or 'calcNormFactors' in edgeR). Unless you have very good reasons to believe that these functions do not work for your data, the results of your analysis will become worse if, instead of using them, you enter library sizes manually. See the paper by Oshlack and Robinson (and also ours on DESeq). Simon
ADD REPLY

Login before adding your answer.

Traffic: 563 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6