Issue with juicer output hic file

[iacopogentile]

Hi everyone, Thank you for developing this tool. I run juicer and it completed by reporting the following messages:

(-: Postprocessing successfully completed, maps too sparse to annotate or GPUs unavailable (-: (-: Pipeline successfully completed (-: Run cleanup.sh to remove the splits directory Check ./opt/juicer/HiC-HUMAN-CD14-1-1/aligned for results

So, I went to have a look at the hic output. At first, I noticed that the splits folders were still there. Further by looking at the hic files, they looked like that:

To be honest, it seems there is not any intrachromosomal interaction reported and I am not sure why. Any idea what could have gone wrong? Thanks, IG

[nchernia]

Hello - can you please post the command you ran, what system you're running under, and the JAR version?

On Wed, Mar 24, 2021 at 11:22 AM iacopogentile @.***> wrote:

Hi everyone, Thank you for developing this tool. I run juicer and it completed by reporting the following messages:

(-: Postprocessing successfully completed, maps too sparse to annotate or GPUs unavailable (-: (-: Pipeline successfully completed (-: Run cleanup.sh to remove the splits directory Check ./opt/juicer/HiC-HUMAN-CD14-1-1/aligned for results

So, I went to have a look at the hic output. At first, I noticed that the splits folders were still there. Further by looking at the hic files, they looked like that: [image: Screen Shot 2021-03-24 at 11 11 14 AM] https://user-images.githubusercontent.com/79707258/112334266-ab7a5600-8c91-11eb-8c1a-3a3a862aba91.png

To be honest, it seems there is not any intrachromosomal interaction reported and I am not sure why. Any idea what could have gone wrong? Thanks, IG

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EW7H2R5U7SWPHQTVLY3TFH7SVANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[iacopogentile]

Hi. Thanks for the reply! I used juicer_tools.1.9.9_jcuda.0.8.jar. and I run the following bash script on the cluster:

HiC_seq () {

    local run=$1

./opt/juicer/scripts/juicer.sh -g GRCh38 -d ./opt/juicer/$run -s MboI -p ./opt/juicer/mygenome.chrom.sizes -y ./opt/juicer/restriction_sites -z ./opt/juicer/references/GRCh38_latest_genomic.fna -D ./opt/juicer/ -t 1
}

for folder in $folderlist; do HiC_seq '$folder' & done

thx, IG

[nchernia]

There is something wrong with your restriction site file most likely, since it seems to be thinking all intra chromosomal reads are intra fragment.

Run at the “final” stage (-S final) and use -s none (or fix your restriction site file, but the above is easier)

On Wed, Mar 24, 2021 at 1:44 PM iacopogentile @.***> wrote:

Hi. Thanks for the reply! I used juicer_tools.1.9.9_jcuda.0.8.jar. and I run the following bash script on the cluster:

HiC_seq () {

local run=$1

./opt/juicer/scripts/juicer.sh -g GRCh38 -d ./opt/juicer/$run -s MboI -p ./opt/juicer/mygenome.chrom.sizes -y ./opt/juicer/restriction_sites -z ./opt/juicer/references/GRCh38_latest_genomic.fna -D ./opt/juicer/ -t 1 }

for folder in $folderlist; do HiC_seq '$folder' & done

thx, IG

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215#issuecomment-806029954, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EW3BZRE5MSQCUXCGNQDTFIQI7ANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[iacopogentile]

Hi, Thank you very much for the reply and I really apologize for the late response. I tried what suggested and it did not affect the output. The fact that the process is not actually terminated (even though it was printed so) since the split folders and other files in the aligned folder are still there and have not been removed yet.

[nchernia]

The splits files don't get removed until you explicitly remove them - in case the pipeline fails and you need to restart at some midway point.

You ran at the final stage with site=none and got the same result?

On Thu, Apr 1, 2021 at 12:34 PM iacopogentile @.***> wrote:

Hi, Thank you very much for the reply and I really apologize for the late response. I tried what suggested and it did not affect the output. The fact that the process is not actually terminated (even though it was printed so) since the split folders and other files in the aligned folder are still there and have not been removed yet.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215#issuecomment-812027636, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EW2ZDTC44IUTCVKSYBDTGSOB5ANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[nchernia]

Sorry, you should also update to a newer jar.

Basically it's clear that your fragments are assigned incorrectly so that all intra-chromosomal reads are defined as intra-fragment. And old versions of the jar will remove this. If you don't want to have to rerun from the chimeric handling stage (another possibility), you should just run from final with a new jar and without specifying a fragment file or RE.

On Thu, Apr 1, 2021 at 12:39 PM Neva Durand @.***> wrote:

The splits files don't get removed until you explicitly remove them - in case the pipeline fails and you need to restart at some midway point.

You ran at the final stage with site=none and got the same result?

On Thu, Apr 1, 2021 at 12:34 PM iacopogentile @.***> wrote:

Hi, Thank you very much for the reply and I really apologize for the late response. I tried what suggested and it did not affect the output. The fact that the process is not actually terminated (even though it was printed so) since the split folders and other files in the aligned folder are still there and have not been removed yet.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215#issuecomment-812027636, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EW2ZDTC44IUTCVKSYBDTGSOB5ANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[iacopogentile]

Ok, Thx! I just replaced juicer_tools.jar from the previous version with the 1.22 one and I will repeat as you suggested. Do, I need to remove only the odd-looking hic files before running it again?

[nchernia]

No, they will get overwritten

On Thu, Apr 1, 2021 at 12:56 PM iacopogentile @.***> wrote:

Ok, Thx! I just replaced juicer_tools.jar from the previous version with the 1.22 one and I will repeat as you suggested. Do, I need to remove only the odd-looking hic files before running it again?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215#issuecomment-812040644, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EW5IVCUC77Y4RWN6VVTTGSQT7ANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[iacopogentile]

Hi, Thx! the hic output looks normal. I have checked the sam output flagstat and it reported that 0% of reads were properly paired. I guess this is not really expected and it could be link to the original problem

[nchernia]

They are aligned separately, I wouldn’t worry about it

On Fri, Apr 2, 2021 at 12:38 PM iacopogentile @.***> wrote:

Hi, Thx! the hic output looks normal. I have checked the sam output flagstat and it reported that 0% of reads were properly paired. I guess this is not really expected and it could be link to the original problem

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/aidenlab/juicer/issues/215#issuecomment-812607009, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAK2EWZ6MTYHYIMQXYCPVJDTGXXHVANCNFSM4ZXPNJ7A .

-- Neva Cherniavsky Durand, Ph.D. | she, her, hers Assistant Professor | Molecular and Human Genetics Aiden Lab | Baylor College of Medicine www.aidenlab.org

[iacopogentile]

Great! Thx!