Question

How does the 'paired' argument works in MeDIPS 1.23.2

0

Entering edit mode

stb • 0

@stb-11175

Last seen 7.4 years ago

Hi,

I have some question regarding the ‘paired’ argument and how to use it in MeDIPS (version 1.23.2)

If ‘paired’ = true, are the arguments ‘extend’ and ‘shift’ then ignored automatically?
If ‘paired’ = true, how does normalization with rpkm then works? Is it only based on the reads (as the name implicates) or is it the entire paired-end fragment? I just read about FPKM, where the fragment are considered instead of the reads.
If ‘paired’ = true, is both the first and second mate then imported? Is it the entire fragment that is included in the analysis or just the reads? An example could be when wiggle files are created: in my data I have reads of 49 bp. I set the ‘window_size’ = 50. Is it then the numbers of reads (49 bp) or also fragments (around 200 bp = spanning several windows) that are counted in each window?

One last question, what is the default setting for 'isSecondaryAlignment'? Is there any recommendations on whether to use only primary alignments or not?

Thanks!

Stine

medips medip-seq R pair-end reads • 1.2k views

ADD COMMENT • link updated 8.2 years ago by Lukas Chavez ▴ 570 • written 8.2 years ago by stb • 0

score 0 · Answer 1 · 2016-09-21

Dear Stine, thanks for your request. About 1) Yes, extend and shift is automatically ignored when MEDIPS is run in paired end mode. About 2) In paired end mode rpkm is in fact fpkm as the fragments are considered once. Abt 3) No, the second mate is not imported. However, the first mate has the information to the distance of its properly paired mate. Therefore, DNA fragments spanned from the start of the first mate to the end of the last mate are considered when coverage at genomic bins is calculated. Abt extra question) default is isSecondaryAlignment=FALSE. I always use this setting, but I am sorry that I cannot give a general recommendation. All the best, Lukas On 21 Sep 2016, at 14:09, stb [bioc] <noreply@bioconductor.org<mailto:noreply@bioconductor.org>> wrote: Activity on a post you are following on support.bioconductor.org<https: support.bioconductor.org=""/> User stb<https: support.bioconductor.org="" u="" 11175=""/> wrote Question: How does the 'paired' argument works in MeDIPS 1.23.2<https: support.bioconductor.org="" p="" 87338=""/>: Hi, I have some question regarding the ‘paired’ argument and how to use it in MeDIPS (version 1.23.2) 1. If ‘paired’ = true, are the arguments ‘extend’ and ‘shift’ then ignored automatically? 2. If ‘paired’ = true, how does normalization with rpkm then works? Is it only based on the reads (as the name implicates) or is it the entire paired-end fragment? I just read about FPKM, where the fragment are considered instead of the reads. 3. If ‘paired’ = true, is both the first and second mate then imported? Is it the entire fragment that is included in the analysis or just the reads? An example could be when wiggle files are created: in my data I have reads of 49 bp. I set the ‘window_size’ = 50. Is it then the numbers of reads (49 bp) or also fragments (around 200 bp = spanning several windows) that are counted in each window? One last question, what is the default setting for 'isSecondaryAlignment'? Is there any recommendations on whether to use only primary alignments or not? Thanks! Stine ________________________________ Post tags: medips, medip-seq, R, pair-end reads You may reply via email or visit How does the 'paired' argument works in MeDIPS 1.23.2