Added preprocessing files

preprocessing/Merge_coolers_cis_trans_forBalance_w_oldG2_fc_1_2_3_4.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+
+seqDir="/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/"
+logDir="/groups/gerlich/experiments/Experiments_004800/004812/clusterlog/"
+
+jobName="Merge_cis_trans_cool_exp4812"
+logFile="$logDir${jobName}.log"
+files[0]="${seqDir}G2.fc_1_2_3_4.wOldG2.cis.1000.cool"
+files[1]="${seqDir}G2.fc_1_2_3_4.wOldG2.trans.1000.cool"
+tempScript="cooler merge ${seqDir}G2.fc_1_2_3_4.wOldG2.cis_and_trans.1000.cool ${files[@]}"
+echo $tempScript
+sbatch -c 1 --mem 20G --qos=short --job-name $jobName --output $logFile --wrap="$tempScript"
+

preprocessing/Transfer_weights_from_cis_and_trans_wOldG2_fc_1_2_3_4.py

@@ -0,0 +1,44 @@
+"""Script used to transfer balancing weigths from
+G2.cis_and_trans.1000.mcool to the cis and trans G2 samples"""
+
+import h5py
+import os
+
+# define functions
+
+
+def transferWeights(source, target, binsize):
+    # get weights from source
+    with h5py.File(source, 'r') as f:
+        weights = f["resolutions"][binsize]["bins"]["weight"]
+        weightArray = weights[:]
+    # write weights into target
+    with h5py.File(target, 'r+') as f:
+        try:
+            targetWeights = f["resolutions"][binsize]["bins"]["weight"]
+            targetWeights[...] = weightArray
+        except KeyError:
+            f["resolutions"][binsize]["bins"]["weight"] = weightArray
+
+
+# set working directory
+
+os.chdir("/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/")
+
+
+# transfer weights for cis and transds
+
+targetDir = "/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/"
+
+
+# get different bins
+bins = "1000,2000,4000,5000,6000,8000,10000,20000,30000,40000,50000,100000,120000,150000,160000,180000,200000,500000,1000000,5000000".split(",")
+# transfer weights trans
+source = 'G2.fc_1_2_3_4.wOldG2.cis_and_trans.1000.mcool'
+target = f'{targetDir}G2.fc_1_2_3_4.wOldG2.trans.1000.mcool'
+for binSize in bins:
+    transferWeights(source, target, binSize)
+# transfer weights cis
+target = f'{targetDir}G2.fc_1_2_3_4.wOldG2.cis.1000.mcool'
+for binSize in bins:
+    transferWeights(source, target, binSize)

preprocessing/Zoomify_and_balance_all_wOldG2_exp4812_ignore_diag_1_fc_1_2_3_4.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+
+# set paths
+coolP="/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/"
+logDir="/groups/gerlich/experiments/Experiments_004800/004812/clusterlog/"
+
+jobName="Zoomify_balance_all"
+logFile="$logDir${jobName}.log"
+tempScript="cooler zoomify ${coolP}G2.fc_1_2_3_4.wOldG2.all.1000.cool -n 22 \
+                             -r 1000,2000,4000,5000,6000,8000,10000,20000,30000,40000,50000,100000,120000,150000,160000,180000,200000,500000,1000000,5000000 --balance\
+                            --balance-args '--ignore-diags 1 --mad-max 5 --max-iters 500'"
+echo $tempScript
+sbatch -c 24 --mem 40G --partition=m --qos=short --job-name $jobName --output $logFile --wrap="$tempScript"

preprocessing/Zoomify_and_balance_cis_and_trans_wOldG2_exp4812_ignore_diag_1_fc_1_2_3_4.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+
+# set paths
+coolP="/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/"
+logDir="/groups/gerlich/experiments/Experiments_004800/004812/clusterlog/"
+
+jobName="Zoomify_balance_cis_and_trans"
+logFile="$logDir${jobName}.log"
+tempScript="cooler zoomify ${coolP}G2.fc_1_2_3_4.wOldG2.cis_and_trans.1000.cool -n 22 \
+                             -r 1000,2000,4000,5000,6000,8000,10000,20000,30000,40000,50000,100000,120000,150000,160000,180000,200000,500000,1000000,5000000 --balance\
+                            --balance-args '--ignore-diags 1 --mad-max 5 --max-iters 500'"
+echo $tempScript
+sbatch -c 24 --mem 40G --partition=m --qos=short --job-name $jobName --output $logFile --wrap="$tempScript"

preprocessing/Zoomify_cis_trans_wOldG2_exp4812_fc_1_2_3_4.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+# set paths
+coolP="/groups/gerlich/experiments/Experiments_004800/004812/Sequencing_data/Pooled_FC_1_2_3_4/cooler/"
+logDir="/groups/gerlich/experiments/Experiments_004800/004812/clusterlog/"
+
+ctypes="cis trans"
+
+
+for ctype in $ctypes; do
+    jobName="Zoomify_${ctype}_${barcode}"
+    logFile="$logDir${jobName}.log"
+    tempScript="cooler zoomify ${coolP}G2.fc_1_2_3_4.wOldG2.${ctype}.1000.cool -n 1 \
+                                -r 1000,2000,4000,5000,6000,8000,10000,20000,30000,40000,50000,100000,120000,150000,160000,180000,200000,500000,1000000,5000000"
+    echo $tempScript
+    sbatch -c 2 --mem 15G --qos=short --job-name $jobName --output $logFile --wrap="$tempScript"
+done

readme.md

@@ -2,6 +2,16 @@
 
 This repository contains notebooks for the analysis of scsHi-C data.
 
+## Preprocessing
+
+Preprocessing of scsHi-C experiments was done as follows: First, the [scshic_pipeline](https://github.com/gerlichlab/scshic_pipeline) was applied to raw sequencing data. The resulting cooler files were then balanced as follows:
+
+- All-contact coolers were zoomified and balanced conventionally (see the following [script](https://github.com/Mittmich/scsHiCanalysis/blob/master/preprocessing/Zoomify_and_balance_all_wOldG2_exp4812_ignore_diag_1_fc_1_2_3_4.sh) for an example for G2 WT data)
+- Cis-sister and trans-sister contacts were pooled (see the following [script](https://github.com/Mittmich/scsHiCanalysis/blob/master/preprocessing/Merge_coolers_cis_trans_forBalance_w_oldG2_fc_1_2_3_4.sh) for an example for G2 WT data)
+- Then, the pooled cis-and-trans sister contacts were zoomified and balanced (see the following [script](https://github.com/Mittmich/scsHiCanalysis/blob/master/preprocessing/Zoomify_and_balance_cis_and_trans_wOldG2_exp4812_ignore_diag_1_fc_1_2_3_4.sh) for an example for G2 WT data)
+- Then, cis-sister and trans-sister coolers were seperately zoomified (see the following [script](https://github.com/Mittmich/scsHiCanalysis/blob/master/preprocessing/Zoomify_cis_trans_wOldG2_exp4812_fc_1_2_3_4.sh) for an example for G2 WT data) and the obtained weights from the cis-and-trans sister coolers transferred to the cis-sister and trans-sister file respectively (see the following [script](https://github.com/Mittmich/scsHiCanalysis/blob/master/preprocessing/Transfer_weights_from_cis_and_trans_wOldG2_fc_1_2_3_4.py) for an example for G2 WT data).
+
+
 ## Fig. 1
 
 ### (d)