Changes between Version 78 and Version 79 of SOPs/atac_Seq


Ignore:
Timestamp:
07/09/21 11:39:30 (4 years ago)
Author:
gbell
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • SOPs/atac_Seq

    v78 v79  
    131131macs2 callpeak -t foo_tn5_pe.bed -n foo -f BED -g mm -q 0.01 --nomodel --shift -75 --extsize 150 --call-summits --keep-dup all
    132132}}}
    133        * If you have biological replicates, you can identify reproducible peaks with the 'idr' function, based on the method of [[https://projecteuclid.org/journals/annals-of-applied-statistics/volume-5/issue-3/Measuring-reproducibility-of-high-throughput-experiments/10.1214/11-AOAS466.full | Qunhua Li et al]] and using [[https://github.com/nboley/idr | code from Nathan Boley]].
    134 {{{
    135 idr --samples rep1.narrowPeak rep2.narrowPeak --input-file-type narrowPeak --output-file IDR.txt --plot
    136 }}}
    137133
    138134    * If you are working with the nucleosome free region (NFR) ( detail information can be found in the bottom of the page), macs2 BAMPE option also works. When using'BAMPE' option with paired-end reads, we let MACS run the pileup and calculate 'extsize'.
     
    150146}}}
    151147 
     148
    152149==== Other Tools or Options ====
    153150
     
    165162                 | grep -P 'chr[\dXY]+[ \t]'  | gzip -nc > ${FILTERED_PEAK}
    166163
     164}}}
     165
     166=== Subset peaks into reproducible peaks ===
     167
     168  * If you have biological replicates, you can identify reproducible peaks with the 'idr' function, based on the method of [[https://projecteuclid.org/journals/annals-of-applied-statistics/volume-5/issue-3/Measuring-reproducibility-of-high-throughput-experiments/10.1214/11-AOAS466.full | Qunhua Li et al]] and using [[https://github.com/nboley/idr | code from Nathan Boley]].
     169{{{
     170idr --samples rep1.narrowPeak rep2.narrowPeak --input-file-type narrowPeak --output-file IDR.narrowPeak.bed --plot
     171}}}
     172  * The output file contains the '''scaled IDR value''' (min(int(-125*log2(IDR), 1000)) in the 5th field.  If one wants to choose 0.05 as the IDR threshold to identify "reproducible" peaks, then this metric must be at least 540.
     173{{{
     174awk '$5 >= 540 {print $0}' IDR.narrowPeak.bed > IDR.narrowPeak.filtered.bed
    167175}}}
    168176