Skip to content

Commit

Permalink
fix links
Browse files Browse the repository at this point in the history
  • Loading branch information
tjparnell committed Nov 1, 2024
1 parent 8ae4012 commit 0186fbc
Show file tree
Hide file tree
Showing 4 changed files with 22 additions and 21 deletions.
19 changes: 10 additions & 9 deletions docs/DeDuplicationEvaluation/DeDuplication.md
Original file line number Diff line number Diff line change
Expand Up @@ -46,12 +46,13 @@ application from the
Reads were aligned to the genome using Novocraft Novoalign (version 4.3).

Alignment bam files were processed with either
[bam_partial_dedup.pl](applications/bam_partial_dedup) or with
[bam_umi_dedup.pl](https://huntsmancancerinstitute.github.io/UMIScripts/apps/bam_umi_dedup.html) from the [UMIScripts
package](https://huntsmancancerinstitute.github.io/UMIScripts/). With partial
de-duplication, all optical duplicates (optical threshold of 2500 pixels) were
completely removed. Exclusion peaks, identified from the corresponding Input, were
called (without prior de-duplication) and used to exclude alignments in all cases.
[bam_partial_dedup.pl](../applications/bam_partial_dedup) or with
[bam_umi_dedup.pl](https://huntsmancancerinstitute.github.io/UMIScripts/apps/bam_umi_dedup.html)
from the [UMIScripts package](https://huntsmancancerinstitute.github.io/UMIScripts/).
With partial de-duplication, all optical duplicates (optical threshold of 2500
pixels) were completely removed. Exclusion peaks, identified from the corresponding
Input, were called (without prior de-duplication) and used to exclude alignments in
all cases.

Six different permutations were generated and compared. Labels are with respect to
the number of duplicates retained.
Expand All @@ -75,13 +76,13 @@ Sequence files are from unpublished results and currently unavailable.

Several plots are provided here as generated by the pipeline.

See [H3K4Me3 results](DeDuplicationEvaluation/H3K4Me3_results.md) for a high
See [H3K4Me3 results](H3K4Me3_results.md) for a high
occurrence, high occupancy factor.

See [RNA Polymerase II results](DeDuplicationEvaluation/RNAPol2_results.md) for a
See [RNA Polymerase II results](RNAPol2_results.md) for a
very high occurrence, moderate occupancy factor.

See [Transcription Factor results](DeDuplicationEvaluation/TF_results.md) for a low
See [Transcription Factor results](TF_results.md) for a low
occurrence, low occupancy factor.


Expand Down
8 changes: 4 additions & 4 deletions docs/DeDuplicationEvaluation/H3K4Me3_results.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,23 +15,23 @@ The number of peaks called between different subsets were nearly identical. Usin
alignments called the most number of peaks. The mean peak length was 1568 ± 1140 bp,
median 1350 bp at a threshold q-value of 2.

![H3K4me3_peak_number](DeDuplicationEvaluation/H3K4me3.peak_number.png)
![H3K4me3_peak_number](H3K4me3.peak_number.png)


### ChIP Efficiency

The fraction of alignments within the respectively called peaks was quite good, with
nearly identical fractions between the subsets, ranging from 43-45%.

![H3K4me3_chip_efficiency](DeDuplicationEvaluation/H3K4me3.chip_efficiency.png)
![H3K4me3_chip_efficiency](H3K4me3.chip_efficiency.png)


### Peak Fragment Coverage Profile

The depth-normalized fragment coverage profile over the called peak midpoint ± 1 Kb
shows no difference between the subsets.

![H3K4me3_profile_fragment](DeDuplicationEvaluation/H3K4me3_profile_fragment_hm.png)
![H3K4me3_profile_fragment](H3K4me3_profile_fragment_hm.png)


### Comparison of Duplicate Alignments Within Peaks
Expand All @@ -47,7 +47,7 @@ duplicate alignments found within peaks. Notably, however, the difference is onl
few percentage points, and more importantly, very little with UMI-deduplicated
alignments.

![H3K4me3_efficiency_comparison](DeDuplicationEvaluation/H3K4me3_comparison.chip_efficiency.png)
![H3K4me3_efficiency_comparison](H3K4me3_comparison.chip_efficiency.png)


### Conclusion
Expand Down
8 changes: 4 additions & 4 deletions docs/DeDuplicationEvaluation/RNAPol2_results.md
Original file line number Diff line number Diff line change
Expand Up @@ -19,23 +19,23 @@ retained duplicates.

The mean peak length was 1140 ± 1883 bp, median 635 bp at a threshold q-value of 2.

![RNAPol2_peak_number](DeDuplicationEvaluation/RNAPol2.peak_number.png)
![RNAPol2_peak_number](RNAPol2.peak_number.png)


### ChIP Efficiency

The fraction of alignments within the respectively called peaks was quite good, with
nearly identical fractions between the subsets, ranging from 33-35%.

![RNAPol2_chip_efficiency](DeDuplicationEvaluation/RNAPol2.chip_efficiency.png)
![RNAPol2_chip_efficiency](RNAPol2.chip_efficiency.png)


### Peak Fragment Coverage Profile

The depth-normalized fragment coverage profile over the called peak midpoint ± 1 Kb
shows no difference between the subsets.

![RNAPol2_profile_fragment](DeDuplicationEvaluation/RNAPol2_profile_fragment_hm.png)
![RNAPol2_profile_fragment](RNAPol2_profile_fragment_hm.png)


### Comparison of Duplicate Alignments Within Peaks
Expand All @@ -51,7 +51,7 @@ duplicate alignments found within peaks. Notably, however, the difference is onl
few percentage points, and more importantly, very little with UMI-deduplicated
alignments.

![RNAPol2_efficiency_comparison](DeDuplicationEvaluation/RNAPol2_comparison.chip_efficiency.png)
![RNAPol2_efficiency_comparison](RNAPol2_comparison.chip_efficiency.png)


### Conclusion
Expand Down
8 changes: 4 additions & 4 deletions docs/DeDuplicationEvaluation/TF_results.md
Original file line number Diff line number Diff line change
Expand Up @@ -20,23 +20,23 @@ Leaving in all duplicates calls nearly 40% more peaks.

The mean peak length was 418 ± 182 bp, median 370 bp at a threshold q-value of 2.

![TF_peak_number](DeDuplicationEvaluation/TF.peak_number.png)
![TF_peak_number](TF.peak_number.png)


### ChIP Efficiency

The fraction of alignments within the respectively called peaks was quite low, as
expected for such few peak numbers called, ranging from 0.25% to 0.30%.

![TF_chip_efficiency](DeDuplicationEvaluation/TF.chip_efficiency.png)
![TF_chip_efficiency](TF.chip_efficiency.png)


### Peak Fragment Coverage Profile

The depth-normalized fragment coverage profile over the called peak midpoint ± 1 Kb
shows no difference between the subsets.

![TF_profile_fragment](DeDuplicationEvaluation/TF_profile_fragment_hm.png)
![TF_profile_fragment](TF_profile_fragment_hm.png)


### Comparison of Duplicate Alignments Within Peaks
Expand All @@ -51,7 +51,7 @@ the peaks than completely deduplicated alignments, indicating there are indeed
duplicate alignments found within peaks. Notably, however, the difference is quite
small.

![TF_efficiency_comparison](DeDuplicationEvaluation/TF_comparison.chip_efficiency.png)
![TF_efficiency_comparison](TF_comparison.chip_efficiency.png)


### Conclusion
Expand Down

0 comments on commit 0186fbc

Please sign in to comment.