Merge pull request #69 from yakirr/master

fixed signed_sumstat datatype bug + 2 minor ldscore computation bugs
bulik · Feb 12, 2017 · c42e055 · c42e055
2 parents f7c3316 + 4be3771
commit c42e055
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 4 deletions.
diff --git a/ldscore/ldscore.py b/ldscore/ldscore.py
@@ -166,7 +166,7 @@ def __corSumVarBlocks__(self, block_left, c, func, snp_getter, annot=None):
         m, n = self.m, self.n
         block_sizes = np.array(np.arange(m) - block_left)
         block_sizes = np.ceil(block_sizes / c)*c
-        if not np.any(annot):
+        if annot is None:
             annot = np.ones((m, 1))
         else:
             annot_m = annot.shape[0]
@@ -204,7 +204,7 @@ def __corSumVarBlocks__(self, block_left, c, func, snp_getter, annot=None):
             # this happens w/ sparse categories (i.e., pathways)
             # update the block
             old_b = b
-            b = block_sizes[l_B]
+            b = int(block_sizes[l_B])
             if l_B > b0 and b > 0:
                 # block_size can't increase more than c
                 # block_size can't be less than c unless it is zero

diff --git a/munge_sumstats.py b/munge_sumstats.py
@@ -634,8 +634,14 @@ def munge_sumstats(args, p=True):
             merge_alleles = None
 
         (openfunc, compression) = get_compression(args.sumstats)
-        dat_gen = pd.read_csv(args.sumstats, delim_whitespace=True, header=0, compression=compression,
-                              usecols=cname_translation.keys(), na_values=['.', 'NA'], iterator=True, chunksize=args.chunksize)
+
+        # figure out which columns are going to involve sign information, so we can ensure
+        # they're read as floats
+        signed_sumstat_cols = [k for k,v in cname_translation.items() if v=='SIGNED_SUMSTAT']
+        dat_gen = pd.read_csv(args.sumstats, delim_whitespace=True, header=0,
+                compression=compression, usecols=cname_translation.keys(),
+                na_values=['.', 'NA'], iterator=True, chunksize=args.chunksize,
+                dtype={c:np.float64 for c in signed_sumstat_cols})
 
         dat = parse_dat(dat_gen, cname_translation, merge_alleles, log, args)
         if len(dat) == 0: