diff options
Diffstat (limited to 'wqflask')
-rw-r--r-- | wqflask/wqflask/my_pylmm/pyLMM/lmm.py | 17 | ||||
-rw-r--r-- | wqflask/wqflask/my_pylmm/pyLMM/runlmm.py | 17 |
2 files changed, 25 insertions, 9 deletions
diff --git a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py index ad6375e9..e51742c4 100644 --- a/wqflask/wqflask/my_pylmm/pyLMM/lmm.py +++ b/wqflask/wqflask/my_pylmm/pyLMM/lmm.py @@ -795,9 +795,9 @@ class LMM: pl.title(title) -def run_gwas(species,k,y,geno,cov,reml,refit,inputfn,new_code): +def run_gwas(species,n,m,k,y,geno,cov=None,reml=True,refit=False,inputfn=None,new_code=True): """ - Invoke pylmm using a genotype (SNP) iterator + Invoke pylmm using genotype as a matrix or as a (SNP) iterator. """ info("gwas_without_redis") print('pheno', y) @@ -848,8 +848,11 @@ def gwas_with_redis(key,species,new_code=True): if v is not None: v = np.array(v) return v - - ps,ts = run_gwas(species,narray('kinship_matrix'),narray('pheno_vector'),narray('genotype_matrix'),narray('covariate_matrix'),params['restricted_max_likelihood'],params['refit'],params['input_file_name'],new_code) + + y = narray('pheno_vector') + n = len(y) + m = params['num_genotypes'] + ps,ts = run_gwas(species,n,m,narray('kinship_matrix'),y,narray('genotype_matrix'),narray('covariate_matrix'),params['restricted_max_likelihood'],params['refit'],params['input_file_name'],new_code) results_key = "pylmm:results:" + params['temp_uuid'] @@ -873,6 +876,7 @@ def gn2_load_redis(key,species,kinship,pheno,geno,new_code=True): k = kinship.tolist() params = dict(pheno_vector = pheno.tolist(), genotype_matrix = geno.tolist(), + num_genotypes = geno.shape[1], kinship_matrix = k, covariate_matrix = None, input_file_name = None, @@ -881,8 +885,7 @@ def gn2_load_redis(key,species,kinship,pheno,geno,new_code=True): temp_uuid = "testrun_temp_uuid", # meta data - timestamp = datetime.datetime.now().isoformat(), - ) + timestamp = datetime.datetime.now().isoformat()) json_params = json.dumps(params) Redis.set(key, json_params) @@ -907,7 +910,7 @@ def gn2_load_redis_iter(key,species,kinship,pheno,geno_iterator): k = kinship.tolist() params = dict(pheno_vector = pheno.tolist(), genotype_matrix = "iterator", - genotypes = i, + num_genotypes = i, kinship_matrix = k, covariate_matrix = None, input_file_name = None, diff --git a/wqflask/wqflask/my_pylmm/pyLMM/runlmm.py b/wqflask/wqflask/my_pylmm/pyLMM/runlmm.py index 3801529e..f095bb73 100644 --- a/wqflask/wqflask/my_pylmm/pyLMM/runlmm.py +++ b/wqflask/wqflask/my_pylmm/pyLMM/runlmm.py @@ -21,7 +21,7 @@ from optparse import OptionParser import sys import tsvreader import numpy as np -from lmm import gn2_load_redis, gn2_load_redis_iter, calculate_kinship_new +from lmm import gn2_load_redis, gn2_load_redis_iter, calculate_kinship_new, run_gwas from kinship import kinship, kinship_full import genotype import phenotype @@ -103,7 +103,20 @@ if options.geno and cmd != 'iterator': g = tsvreader.geno(options.geno) print g.shape -if cmd == 'iterator': +if cmd == 'run': + if options.remove_missing_phenotypes: + raise Exception('Can not use --remove-missing-phenotypes with LMM2') + snp_iterator = tsvreader.geno_iter(options.geno) + n = len(y) + m = g.shape[1] + ps, ts = run_gwas('other',n,m,k,y,g.T) + print np.array(ps) + print len(ps),sum(ps) + # Test results + p1 = round(ps[0],4) + p2 = round(ps[-1],4) + +elif cmd == 'iterator': if options.remove_missing_phenotypes: raise Exception('Can not use --remove-missing-phenotypes with LMM2') snp_iterator = tsvreader.geno_iter(options.geno) |