Got Nick's code working with the MDP datasets under the BXD group,

but with only BXD strains

7 files changed, 53 insertions, 17 deletions

diff --git a/wqflask/wqflask/marker_regression/marker_regression.py b/wqflask/wqflask/marker_regression/marker_regression.py
index f67c595a..5f4f9be8 100755
--- a/wqflask/wqflask/marker_regression/marker_regression.py
+++ b/wqflask/wqflask/marker_regression/marker_regression.py
@@ -9,6 +9,7 @@ import string
 import sys
 import os
 import collections
+import pdb
 
 import numpy as np
 from scipy import linalg
@@ -67,13 +68,13 @@ class MarkerRegression(object):
         pheno_vector = np.array([val == "x" and np.nan or float(val) for val in self.vals])
 
         if self.dataset.group.species == "human":
-            p_values, t_stats = self.gen_human_results(pheno_vector)
+            p_values, t_stats = self.gen_human_results(pheno_vector, tempdata)
         else:
             genotype_data = [marker['genotypes'] for marker in self.dataset.group.markers.markers]
             
             no_val_samples = self.identify_empty_samples()
-            trimmed_genotype_data = self.trim_genotypes(genotype_data, no_val_samples)
-            
+            trimmed_genotype_data = self.trim_genotypes(genotype_data, no_value_samples=[])
+            pdb.set_trace()
             genotype_matrix = np.array(trimmed_genotype_data).T
             
             print("pheno_vector is: ", pf(pheno_vector))
@@ -92,10 +93,12 @@ class MarkerRegression(object):
         self.qtl_results = self.dataset.group.markers.markers
 
 
-    def gen_human_results(self, pheno_vector):
+    def gen_human_results(self, pheno_vector, tempdata):
         file_base = os.path.join(webqtlConfig.PYLMM_PATH, self.dataset.group.name)
         
+        tempdata.store("percent_complete", 0)
         plink_input = input.plink(file_base, type='b')
+        tempdata.store("percent_complete", 0.1)
 
         pheno_vector = pheno_vector.reshape((len(pheno_vector), 1))
         covariate_matrix = np.ones((pheno_vector.shape[0],1))
@@ -106,7 +109,8 @@ class MarkerRegression(object):
                 pheno_vector,
                 covariate_matrix,
                 plink_input,
-                kinship_matrix
+                kinship_matrix,
+                temp_data=tempdata
             )
 
         return p_values, t_stats
diff --git a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
index e60f7b02..4de77173 100644
--- a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
+++ b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py
@@ -23,6 +23,7 @@ import numpy as np
 from scipy import linalg
 from scipy import optimize
 from scipy import stats
+import pdb
 
 from pprint import pformat as pf
 
@@ -43,10 +44,10 @@ def run_human(pheno_vector,
 
     if v.sum():
         pheno_vector = pheno_vector[keep]
-        print("pheno_vector shape is now: ", pf(pheno_vector.shape))
+        #print("pheno_vector shape is now: ", pf(pheno_vector.shape))
         covariate_matrix = covariate_matrix[keep,:]
-        print("kinship_matrix shape is: ", pf(kinship_matrix.shape))
-        print("len(keep) is: ", pf(keep.shape))
+        #print("kinship_matrix shape is: ", pf(kinship_matrix.shape))
+        #print("len(keep) is: ", pf(keep.shape))
         kinship_matrix = kinship_matrix[keep,:][:,keep]
 
     n = kinship_matrix.shape[0]
@@ -58,13 +59,21 @@ def run_human(pheno_vector,
     # Buffers for pvalues and t-stats
     p_values = []
     t_stats = []
+    
+    plink_input.getSNPIterator()
+    total_snps = plink_input.numSNPs
+    
     count = 0
     with Bench("snp iterator loop"):
         for snp, this_id in plink_input:
-            if count > 1000:
-                break
+            #if count > 1000:
+            #    break
             count += 1
 
+            percent_complete = (float(count) / total_snps) * 100
+            print("percent_complete: ", pf(percent_complete))
+            temp_data.store("percent_complete", percent_complete)        
+
             x = snp[keep].reshape((n,1))
             #x[[1,50,100,200,3000],:] = np.nan
             v = np.isnan(x).reshape((-1,))
@@ -104,9 +113,12 @@ def run_human(pheno_vector,
                 if refit:
                     lmm_ob.fit(X=x)
                 ts, ps, beta, betaVar = lmm_ob.association(x)
+                
             p_values.append(ps)
             t_stats.append(ts)
             
+    print("p_values: ", pf(p_values))        
+    
     return p_values, t_stats
 
 
@@ -188,14 +200,13 @@ def calculate_kinship(genotype_matrix, temp_data):
     #print("m is:", m)
     keep = []
     for counter in range(m):
-        print("type of genotype_matrix[:,counter]:", pf(genotype_matrix[:,counter]))
+        #print("type of genotype_matrix[:,counter]:", pf(genotype_matrix[:,counter]))
         #Checks if any values in column are not numbers
         not_number = np.isnan(genotype_matrix[:,counter])
-        print("type of not_number:", type(not_number))
         
         #Gets vector of values for column (no values in vector if not all values in col are numbers)
         marker_values = genotype_matrix[True - not_number, counter]
-        print("type of marker_values is:", type(marker_values))
+        #print("type of marker_values is:", type(marker_values))
         
         #Gets mean of values in vector
         values_mean = marker_values.mean()
@@ -270,7 +281,10 @@ def GWAS(pheno_vector,
 
     p_values = []
     t_statistics = []
-
+    
+    n = genotype_matrix.shape[0]
+    m = genotype_matrix.shape[1]
+    
     for counter in range(m):
         x = genotype_matrix[:,counter].reshape((n, 1))
         v = np.isnan(x).reshape((-1,))
@@ -343,6 +357,7 @@ class LMM:
  
        #x = Y != -9
        x = True - np.isnan(Y)
+       #pdb.set_trace()
        if not x.sum() == len(Y):
           if self.verbose: sys.stderr.write("Removing %d missing values from Y\n" % ((True - x).sum()))
           Y = Y[x]
@@ -362,8 +377,8 @@ class LMM:
        self.K = K
        self.Kva = Kva
        self.Kve = Kve
-       print("self.Kva is: ", pf(self.Kva))
-       print("self.Kve is: ", pf(self.Kve))
+       #print("self.Kva is: ", pf(self.Kva))
+       #print("self.Kve is: ", pf(self.Kve))
        self.Y = Y
        self.X0 = X0
        self.N = self.K.shape[0]
diff --git a/wqflask/wqflask/static/new/javascript/marker_regression.coffee b/wqflask/wqflask/static/new/javascript/marker_regression.coffee
index 3e14ab6b..78b6fdbc 100644
--- a/wqflask/wqflask/static/new/javascript/marker_regression.coffee
+++ b/wqflask/wqflask/static/new/javascript/marker_regression.coffee
@@ -4,10 +4,12 @@ $ ->
             @qtl_results = js_data.qtl_results
             console.log("qtl_results are:", @qtl_results)
             @chromosomes = js_data.chromosomes
+            console.log("chromosomes: ", @chromosomes)
 
             @total_length = 0
 
             @max_chr = @get_max_chr()
+            console.log("max_chr is: ", @max_chr)
 
             @x_coords = []
             @y_coords = []
diff --git a/wqflask/wqflask/static/new/javascript/marker_regression.js b/wqflask/wqflask/static/new/javascript/marker_regression.js
index 09470daf..25d88ec0 100644
--- a/wqflask/wqflask/static/new/javascript/marker_regression.js
+++ b/wqflask/wqflask/static/new/javascript/marker_regression.js
@@ -13,8 +13,10 @@
         this.qtl_results = js_data.qtl_results;
         console.log("qtl_results are:", this.qtl_results);
         this.chromosomes = js_data.chromosomes;
+        console.log("chromosomes: ", this.chromosomes);
         this.total_length = 0;
         this.max_chr = this.get_max_chr();
+        console.log("max_chr is: ", this.max_chr);
         this.x_coords = [];
         this.y_coords = [];
         this.marker_names = [];
diff --git a/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.coffee b/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.coffee
index 157f56a9..22427e4f 100644
--- a/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.coffee
+++ b/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.coffee
@@ -37,6 +37,12 @@ $ ->
             type: "POST"
             url: url
             data: form_data
+            error: (xhr, ajaxOptions, thrownError) =>
+                alert("some error occurred")
+                console.log(xhr)
+                clearInterval(this.my_timer)
+                $('#progress_bar_container').modal('hide')
+                $("body").html("error")        
             success: (data) =>
                 clearInterval(this.my_timer)
                 $('#progress_bar_container').modal('hide')
diff --git a/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.js b/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.js
index 78459692..daf9b273 100644
--- a/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.js
+++ b/wqflask/wqflask/static/new/javascript/show_trait_mapping_tools.js
@@ -46,6 +46,13 @@
         type: "POST",
         url: url,
         data: form_data,
+        error: function(xhr, ajaxOptions, thrownError) {
+          alert("some error occurred");
+          console.log(xhr);
+          clearInterval(_this.my_timer);
+          $('#progress_bar_container').modal('hide');
+          return $("body").html("error");
+        },
         success: function(data) {
           clearInterval(_this.my_timer);
           $('#progress_bar_container').modal('hide');
diff --git a/wqflask/wqflask/views.py b/wqflask/wqflask/views.py
index 7f5f88e0..c303e0d1 100644
--- a/wqflask/wqflask/views.py
+++ b/wqflask/wqflask/views.py
@@ -168,7 +168,7 @@ def marker_regression_page():
         if key in wanted or key.startswith(('value:')):
             start_vars[key] = value
     
-    version = "v13"
+    version = "v14"
     key = "marker_regression:{}:".format(version) + json.dumps(start_vars, sort_keys=True)
     with Bench("Loading cache"):
         result = Redis.get(key)