p300 support

CHANGELOG.md

@@ -6,6 +6,10 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ## Unreleased
 
+### Added
+
+- experimental p300 ChIP-seq support
+
 ### Changed
 
 - updated dependencies

anansnake/envs/ananse.yaml

@@ -3,5 +3,11 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - python =3.10
-  - ananse =0.5.0
+  - conda-forge::python =3.10
+  - bioconda::ananse =0.5.1
+  - conda-forge::icu =73  # ImportError: /lib/x86_64-linux-gnu/libstdc++.so.6: version `CXXABI_1.3.15' not found (sql and pyarrow import order related)
+  - conda-forge::conda-ecosystem-user-package-isolation=1.0
+
+  - pip
+  - pip:
+      - git+https://github.com/vanheeringen-lab/ANANSE.git@p300_2

anansnake/envs/gimme.yaml

@@ -3,8 +3,13 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - python =3.10
-  - gimmemotifs-minimal =0.18.0
+  - python =3.11
+  #- gimmemotifs-minimal =0.18.3
   - gffread
   - orthofinder
-  - xgboost  # optional maelstrom dependency
+  - conda-forge::conda-ecosystem-user-package-isolation=1.0
+
+  - pip
+  - pip:
+      - git+https://github.com/vanheeringen-lab/gimmemotifs@scanner_fix
+  - xgboost >= 1.0.2  # cannot be installed using pip

anansnake/rules/ananse.smk

@@ -14,11 +14,13 @@ rule binding:
         expand("{binding_dir}/{{condition}}.h5", ** config),
     log:
         expand("{log_dir}/binding_{{condition}}.txt", **config),
-    benchmark:
-        expand("{benchmark_dir}/binding_{{condition}}.txt", **config)[0]
+    # benchmark:
+    #     expand("{benchmark_dir}/binding_{{condition}}.txt", **config)[0]
     params:
         atac_samples=lambda wildcards: CONDITIONS[wildcards.condition]["ATAC-seq samples"],
+        enhancer_data=lambda wildcards: "-P" if config.get("enhancer_data", "").lower() == "p300" else ("-H" if config.get("enhancer_data", "").lower() == "h3k27ac" else "-A"),
         jaccard=config["jaccard"],
+        tfs=f"--tfs {config['tfs']} if "tfs" in config else "",
     threads: 1  # multithreading not required when using a pfmscorefile
     resources:
         mem_mb=40_000,  # 30-50 GB
@@ -33,14 +35,16 @@ rule binding:
 
         # additional log info
         printf "using columns: {params.atac_samples}\n\n" > {log}
+        printf "enhancer data input argument: {params.enhancer_data}\n\n" > {log}
 
         ananse binding \
-        -A {input.atac} \
+        {params.enhancer_data} {input.atac} \
         -c {params.atac_samples} \
         -g {input.genome} \
         -p {input.pfm} \
         --pfmscorefile {input.pfmscorefile} \
         --jaccard-cutoff {params.jaccard} \
+        {params.tfs} \
         -n {threads} \
         -o $outdir \
         >> {log} 2>&1
@@ -73,10 +77,11 @@ rule network:
         expand("{network_dir}/{{condition}}.tsv",**config),
     log:
         expand("{log_dir}/network_{{condition}}.txt",**config),
-    benchmark:
-        expand("{benchmark_dir}/network_{{condition}}.txt",**config)[0]
+    # benchmark:
+    #     expand("{benchmark_dir}/network_{{condition}}.txt",**config)[0]
     params:
         rna_samples=lambda wildcards: CONDITIONS[wildcards.condition]["RNA-seq samples"],
+        tfs=f"--tfs {config['tfs']} if "tfs" in config else "",
     threads: 1  # multithreading explodes memory
     resources:
         network=1,
@@ -96,6 +101,7 @@ rule network:
         {input.binding} \
         -e {input.genes} \
         -c {params.rna_samples} \
+        {params.tfs} \
         -g {input.genome} \
         -o {output} \
         --full-output \
@@ -125,8 +131,8 @@ rule influence:
         diff_inf = expand("{influence_dir}/{{contrast}}_diffnetwork.tsv",** config),
     log:
         expand("{log_dir}/influence_{{contrast}}.txt",** config)[0]
-    benchmark:
-        expand("{benchmark_dir}/influence_{{contrast}}.txt",**config)[0]
+    # benchmark:
+    #     expand("{benchmark_dir}/influence_{{contrast}}.txt",**config)[0]
     params:
         edges=config["edges"],
         padj=config["padj"],

example/config.yaml

@@ -50,6 +50,7 @@ merged_technical_reps: true  # set to false if you used 'technical_replicates: "
 merged_biological_reps: true  # set to false if you used 'biological_replicates: "keep" in s2s'
 
 # ANANSE binding
+enhancer_data: ATAC  # experimental option: p300
 jaccard: 0.2  # default: 0.1
 
 # ANANSE influence