diff --git a/README.md b/README.md
index f2d4ae5..d1422e2 100644
--- a/README.md
+++ b/README.md
@@ -116,6 +116,8 @@ After building the index, all but the refseq_abv.[1234].cfr index files may be r
 * #### Build custom database index 
 The index building procedure is similar to [Centrifuge's](http://www.ccb.jhu.edu/software/centrifuge/manual.shtml#database-download-and-index-building), but with names changing to centrifuger. For example, centrifuge-download is centrifuger-download. 
 
+The folder "indices" contains information for creating index from other sources, like SILVA and GTDB.
+
 * #### 10x Genomics data and barcode-based single-cell data
 
 If your input has barcode information, you can use "--barcode" to specify the barcode file and use "--read-format" to tell Centrifuger how to extract barcode information. The "--read-format" option can also specify the extraction for read1, read2 and UMI. The value for this argument is a comma-separated string, each field in the string is also a semi-comma-splitted string
diff --git a/defs.h b/defs.h
index 6f3578f..3f8f471 100644
--- a/defs.h
+++ b/defs.h
@@ -5,7 +5,7 @@
 
 //#define DEBUG
 
-#define CENTRIFUGER_VERSION "1.0.4-r152"
+#define CENTRIFUGER_VERSION "1.0.4-r153"
 
 extern char nucToNum[26] ; 
 extern char numToNuc[26] ;
diff --git a/indices/README.md b/indices/README.md
index f84b1f3..4eed7b8 100644
--- a/indices/README.md
+++ b/indices/README.md
@@ -16,3 +16,4 @@ The repetitiveness of GTDB representative genomes is very low, so the run-block
 ```
 ../centrifuger-build -l gtdb_fname_to_taxid.map --taxonomy-tree gtdb_nodes.dmp --name-table gtdb_names.dmp -t 16 -o cfr_gtdb --build-mem 500G --offrate 5 --rbbwt-b 1
 ```
+A pre-built index for GTDB r220 representative genomes is available at: https://www.dropbox.com/scl/fo/wtsermtdd62n1ttryuhwv/ADWTYiJA6Dh5gdbHmwC2nfo?rlkey=fgdpthjukcj1uhsjrcv35j927&dl=0 .