diff --git a/doc/manual.rst b/doc/manual.rst
index fbc36ce..565f5be 100644
--- a/doc/manual.rst
+++ b/doc/manual.rst
@@ -32,7 +32,8 @@ to demultiplex scRNA-seq data.
 
       vireo -c $CELL_DATA -N $n_donor -o $OUT_DIR
 
-2) with genotype for all samples (GT, GP, or PL)
+2) with genotype for all samples (genoTag: GT, GP, or PL; default is PL, please 
+   choose the existing one)
 
    ::
 
@@ -41,6 +42,10 @@ to demultiplex scRNA-seq data.
    Optionally, `-N` can be provided if it is samller than that in DONOR_GT_FILE
    for finding the relevant subset of donors.
 
+   **Note** For efficient loading of donor VCF file, we recommend subset it
+   ``bcftools view donor.vcf.gz -R cellSNP.cells.vcf.gz -Oz -o sub.vcf.gz``
+   Also, add ``-s`` or ``-S`` for subsetting samples. 
+
 3) with genotype for part of the samples (n_donor is larger than that in 
    DONOR_GT_FILE)
 
@@ -63,7 +68,7 @@ Viroe supports the cell data in three formats:
 
 
 Vireo full arguments
-====================
+--------------------
 
 Type ``vireo -h`` for details of all arguments:
 
diff --git a/doc/release.rst b/doc/release.rst
index b4e1445..b7e3b04 100644
--- a/doc/release.rst
+++ b/doc/release.rst
@@ -2,6 +2,10 @@
 History
 =======
 
+Release v0.1.7 (05/10/2019)
+===========================
+* Support donor genotype vcf file with different FORMAT for different variants
+
 Release v0.1.6 (05/10/2019)
 ===========================
 * Fix a bug when variants in donor genotype are not in cell vcf file
diff --git a/vireoSNP/utils/vcf_utils.py b/vireoSNP/utils/vcf_utils.py
index 284a7c6..199697d 100644
--- a/vireoSNP/utils/vcf_utils.py
+++ b/vireoSNP/utils/vcf_utils.py
@@ -8,7 +8,7 @@
 import subprocess
 import numpy as np
 
-def parse_sample_info(sample_dat, sparse=True):
+def parse_sample_info(sample_dat, sparse=True, format_list=None):
     """
     Parse genotype information for each sample
     Note, it requires the format for each variants to 
@@ -19,18 +19,19 @@ def parse_sample_info(sample_dat, sparse=True):
 
     # require the same format for all variants
     format_all = [x[0].split(":") for x in sample_dat]
-    format_list = format_all[0]
+    if format_list is None:
+        format_list = format_all[0]
 
-    ## sparse matrix requires all keys
-    format_set_all = [set(x) for x in format_all]
-    if format_set_all.count(set(format_all[0])) != len(format_all):
-        print("Error: require the same format for all variants.")
-        exit()
-    
     RV = {}
     for _key in format_list:
         RV[_key] = []
     if sparse:
+        ## sparse matrix requires all keys
+        format_set_all = [set(x) for x in format_all]
+        if format_set_all.count(set(format_list)) != len(format_all):
+            print("Error: require the same format for all variants.")
+            exit()
+
         RV['indices'] = []
         RV['indptr'] = [0]
         RV['shape'] = (len(sample_dat[0][1:]), len(sample_dat))
@@ -64,7 +65,8 @@ def parse_sample_info(sample_dat, sparse=True):
     return RV
 
 
-def load_VCF(vcf_file, biallelic_only=False, load_sample=True, sparse=True):
+def load_VCF(vcf_file, biallelic_only=False, load_sample=True, sparse=True,
+             format_list=None):
     """
     Load whole VCF file 
     -------------------
@@ -121,7 +123,7 @@ def load_VCF(vcf_file, biallelic_only=False, load_sample=True, sparse=True):
     RV["comments"]  = comment_lines
     if load_sample:
         RV["samples"]   = obs_ids
-        RV["GenoINFO"]  = parse_sample_info(obs_dat, sparse=sparse)
+        RV["GenoINFO"]  = parse_sample_info(obs_dat, sparse, format_list)
     return RV
 
 
diff --git a/vireoSNP/version.py b/vireoSNP/version.py
index 32efefd..283b03a 100644
--- a/vireoSNP/version.py
+++ b/vireoSNP/version.py
@@ -1 +1 @@
-__version__ = "0.1.6"
\ No newline at end of file
+__version__ = "0.1.7"
\ No newline at end of file
diff --git a/vireoSNP/vireo.py b/vireoSNP/vireo.py
index 8a58feb..d46bd59 100644
--- a/vireoSNP/vireo.py
+++ b/vireoSNP/vireo.py
@@ -132,8 +132,8 @@ def main():
     n_donor = options.n_donor
     if options.donor_file is not None:
         print("[vireo] Loading donor VCF file ...")
-        donor_vcf = load_VCF(options.donor_file, sparse=False, 
-                             biallelic_only=True)
+        donor_vcf = load_VCF(options.donor_file, biallelic_only=True, 
+                             sparse=False, format_list=[options.geno_tag])
         if (options.geno_tag not in donor_vcf['GenoINFO']):
             print("[vireo] No " + options.geno_tag + " tag in donor genotype; " 
                 "please try another tag for genotype, e.g., GT")