diff --git a/VERSIONLOG.md b/VERSIONLOG.md
index 4b566065..1a2aadde 100644
--- a/VERSIONLOG.md
+++ b/VERSIONLOG.md
@@ -1,5 +1,8 @@
 # TACA Version Log
 
+## 20210819.1
+Allow 0 mismatch in demux for short single index
+
 ## 20210716.1
 Handle special case with index 2 only for NovaSeq
 
diff --git a/taca/illumina/HiSeqX_Runs.py b/taca/illumina/HiSeqX_Runs.py
index 4458e474..39c53e7c 100755
--- a/taca/illumina/HiSeqX_Runs.py
+++ b/taca/illumina/HiSeqX_Runs.py
@@ -219,6 +219,9 @@ def generate_bcl_command(self, sample_type, mask_table, bcl2fastq_cmd_counter):
                 # Add the extra Smart-seq command options if we have 10X ST samples
                 if sample_type == 'SMARTSEQ':
                     cl_options.extend(self.CONFIG['bcl2fastq']['options_SMARTSEQ'])
+                # Add the extra command option if we have samples with single short index
+                if sample_type == 'short_single_index':
+                    cl_options.extend(self.CONFIG['bcl2fastq']['options_short_single_index'])
                 # Append all options that appear in the configuration file to the main command.
                 for option in cl_options:
                     if isinstance(option, dict):
@@ -452,7 +455,11 @@ def _classify_samples(indexfile, ssparser):
         # Ordinary samples
         else:
             index_length = [len(sample['index']),len(sample['index2'])]
-            sample_type = 'ordinary'
+            # Short single index (<=6nt)
+            if (index_length[0] <= 6 and index_length[1] == 0) or (index_length[0] == 0 and index_length[1] <= 6):
+                sample_type = 'short_single_index'
+            else:
+                sample_type = 'ordinary'
 
         # Write in sample table
         # {'1': [('101', {'sample_type': 'ordinary', 'index_length': [8, 8]}), ('102', {'sample_type': 'ordinary', 'index_length': [8, 8]})]}