Merge pull request #31 from zjnolen/anc_state

Allow polarizing SFS to ancestral states

.test/config/config.yaml

@@ -18,6 +18,8 @@ reference:
   exclude: []
   min_size: 3000
 
+ancestral: "data/ref/testref.fa"
+
 #===================== Sample Set Configuration =======================#
 
 exclude_ind: []

config/README.md

@@ -134,6 +134,13 @@ Required configuration of the reference.
   - `min_size:` A size in bp (integer). All contigs below this size will be
     excluded from analysis.
 
+- `ancestral:` A path to a fasta file containing the ancestral states in your
+  reference genome. This is optional, and is used to polarize allele
+  frequencies in SAF files to ancestral/derived. If you leave this empty,
+  the reference genome itself will be used as ancestral, and you should be
+  sure the [`params`] [`realSFS`] [`fold`] is set to `1`. If you put a reference
+  here, you can set that to `0`.
+
 Reference genomes should be uncompressed, and contig names should be clear and
 concise. Currently, there are some issues parsing contig names with
 underscores, so please change these in your reference before running the
@@ -453,9 +460,8 @@ or a pull request and I'll gladly put it in.
     - `pruning_min-weight:` The minimum r^2 to assume two positions are in
       linkage disequilibrium when pruning (float)
   - `realSFS:` Settings for realSFS
-    - `fold:` Whether or not to fold the produced SFS (0 or 1, [docs](http://www.popgen.dk/angsd/index.php/SFS_Estimation))
-      **NOTE:** I have not implemented the use of an ancestral reference into
-      this workflow, so this should always be set to 1 until I implement this.
+    - `fold:` Whether or not to fold the produced SFS. Set to 1 if you have not
+      provided an ancestral-state reference (0 or 1, [docs](http://www.popgen.dk/angsd/index.php/SFS_Estimation))
     - `sfsboot:` Doesn't work now, but when it does it will produce this many
       bootstrapped SFS per population and population pair (integer)
   - `fst:` Settings for $F_{ST}$ calculation in ANGSD

config/config.yaml

@@ -18,6 +18,8 @@ reference:
   exclude: []
   min_size:
 
+ancestral:
+
 #===================== Sample Set Configuration =======================#
 
 exclude_ind: []
@@ -137,7 +139,7 @@ params:
     fit_LDdecay_n_correction: true
     pruning_min-weight: 0.1
   realsfs:
-    fold: 1 # Should only be set to 1 for now
+    fold: 1 # Should only be set to 1, unless ancestral reference is given above
     sfsboot: 100
   fst:
     whichFst: 1

workflow/rules/0.1_ref_prep.smk

@@ -24,6 +24,24 @@ rule link_ref:
         """
 
 
+if config["ancestral"]:
+
+    rule link_anc_ref:
+        """Link ancestral reference genome to results directory if it exists"""
+        input:
+            config["ancestral"],
+        output:
+            "results/ref/{ref}/{ref}.anc.fa",
+        log:
+            "logs/ref/link_ref/{ref}.anc.log",
+        conda:
+            "../envs/shell.yaml"
+        shell:
+            """
+            ln -sr {input} {output} 2> {log}
+            """
+
+
 rule bwa_index:
     """Index reference genome for bwa (mapping)"""
     input:
@@ -50,13 +68,13 @@ rule bwa_index:
 rule samtools_faidx:
     """Index reference genome using samtools (fai index used by several tools)"""
     input:
-        "results/ref/{ref}/{ref}.fa",
+        "results/ref/{ref}/{prefix}.fa",
     output:
-        "results/ref/{ref}/{ref}.fa.fai",
+        "results/ref/{ref}/{prefix}.fa.fai",
     log:
-        "logs/ref/samtools_faidx/{ref}.log",
+        "logs/ref/samtools_faidx/{ref}/{prefix}.log",
     benchmark:
-        "benchmarks/ref/samtools_faidx/{ref}.log"
+        "benchmarks/ref/samtools_faidx/{ref}/{prefix}.log"
     wrapper:
         "v2.4.0/bio/samtools/faidx"
 

workflow/rules/3.1_safs.smk

@@ -7,10 +7,12 @@ rule angsd_doSaf_pop:
     """
     input:
         unpack(filt_depth),
+        unpack(get_anc_ref),
         bam="results/datasets/{dataset}/bamlists/{dataset}.{ref}_{population}{dp}.bamlist",
         bams=get_bamlist_bams,
         bais=get_bamlist_bais,
         ref="results/ref/{ref}/{ref}.fa",
+        reffai="results/ref/{ref}/{ref}.fa.fai",
         regions="results/datasets/{dataset}/filters/chunks/{ref}_chunk{chunk}.rf",
     output:
         saf=temp(
@@ -50,7 +52,7 @@ rule angsd_doSaf_pop:
         """
         angsd -doSaf 1 -bam {input.bam} -GL {params.gl_model} -ref {input.ref} \
             -nThreads {threads} {params.extra} -minMapQ {params.mapQ} \
-            -minQ {params.baseQ} -sites {input.sites} -anc {input.ref} \
+            -minQ {params.baseQ} -sites {input.sites} -anc {input.anc} \
             {params.extra_saf} -rf {input.regions} -out {params.out} &> {log}
         """
 
@@ -109,10 +111,12 @@ rule angsd_doSaf_sample:
     """
     input:
         unpack(filt_depth),
+        unpack(get_anc_ref),
         bam="results/datasets/{dataset}/bamlists/{dataset}.{ref}_{population}{dp}.bamlist",
         bams=get_bamlist_bams,
         bais=get_bamlist_bais,
         ref="results/ref/{ref}/{ref}.fa",
+        reffai="results/ref/{ref}/{ref}.fa.fai",
     output:
         saf="results/datasets/{dataset}/safs/{dataset}.{ref}_{population}{dp}_{sites}-filts.saf.gz",
         safidx="results/datasets/{dataset}/safs/{dataset}.{ref}_{population}{dp}_{sites}-filts.saf.idx",
@@ -140,7 +144,7 @@ rule angsd_doSaf_sample:
         """
         (angsd -doSaf 1 -bam {input.bam} -GL {params.gl_model} -ref {input.ref} \
             -nThreads {threads} {params.extra} -minMapQ {params.mapQ} \
-            -minQ {params.baseQ} -sites {input.sites} -anc {input.ref} \
+            -minQ {params.baseQ} -sites {input.sites} -anc {input.anc} \
             -setMinDepthInd {params.mindepthind} {params.extra_saf} \
             -out {params.out}) &> {log}
         """

workflow/rules/common.smk

@@ -154,8 +154,8 @@ def get_raw_fastq(wildcards):
             & (units["lib"] == wildcards.lib),
             ["sample", "fq1", "fq2"],
         ].set_index("sample")
-        if not pd.isna(unit.fq1[0]):
-            return {"sample": [unit.fq1[0], unit.fq2[0]]}
+        if not pd.isna(unit.fq1.item()):
+            return {"sample": [unit.fq1.item(), unit.fq2.item()]}
     if "sra" in units:
         sra = units[
             (units["sample"] == wildcards.sample)
@@ -454,6 +454,7 @@ def get_endo_cont_stat(wildcards):
 #         return "results/datasets/{dataset}/glfs/chunks/{dataset}.{ref}_{population}{dp}_chunk{chunk}_allsites-filts.glf.gz"
 
 
+# Select filter file based off of full depth samples or subsampled depth
 def filt_depth(wildcards):
     if config["subsample_redo_filts"]:
         return {
@@ -466,6 +467,19 @@ def filt_depth(wildcards):
     }
 
 
+# Choose to use an ancestral reference if present, otherwise use main reference
+def get_anc_ref(wildcards):
+    if config["ancestral"]:
+        return {
+            "anc": "results/ref/{ref}/{ref}.anc.fa",
+            "ancfai": "results/ref/{ref}/{ref}.anc.fa.fai",
+        }
+    return {
+        "anc": "results/ref/{ref}/{ref}.fa",
+        "ancfai": "results/ref/{ref}/{ref}.fa.fai",
+    }
+
+
 ## Get random sampling proportion depending on if LD decay is being calculated
 ## or if LD pruning is being done
 def get_ngsld_sampling(wildcards):