diff --git a/Cargo.toml b/Cargo.toml
index 97f67f1..b9b79f5 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "STRdust"
-version = "0.10.0"
+version = "0.11.0"
 edition = "2021"
 
 
diff --git a/src/genotype.rs b/src/genotype.rs
index 44e8942..35f04ae 100644
--- a/src/genotype.rs
+++ b/src/genotype.rs
@@ -68,7 +68,7 @@ fn genotype_repeat(
         && args.haploid.as_ref().unwrap().contains(&repeat.chrom))
         || args.unphased;
 
-    let reads = match crate::parse_bam::get_overlapping_reads(bam, repeat, unphased) {
+    let reads = match crate::parse_bam::get_overlapping_reads(bam, repeat, unphased, args.max_number_reads) {
         Some(seqs) => seqs,
         None => {
             // Return a missing genotype if no (phased) reads overlap the repeat
@@ -369,7 +369,7 @@ mod tests {
         let unphased = false;
         let repeat_compressed_reference = repeat.make_repeat_compressed_sequence(&fasta, flanking);
         let mut bam = parse_bam::create_bam_reader(&bam, &fasta);
-        let binding = crate::parse_bam::get_overlapping_reads(&mut bam, &repeat, unphased).unwrap();
+        let binding = crate::parse_bam::get_overlapping_reads(&mut bam, &repeat, unphased, 60).unwrap();
         let read = binding
             .seqs
             .get(&1)
@@ -420,6 +420,7 @@ mod tests {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60
         };
         let mut bam = parse_bam::create_bam_reader(&args.bam, &args.fasta);
         let genotype = genotype_repeat(&repeat, &args, &mut bam);
@@ -451,6 +452,7 @@ mod tests {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60
         };
         let mut bam = parse_bam::create_bam_reader(&args.bam, &args.fasta);
         let genotype = genotype_repeat(&repeat, &args, &mut bam);
@@ -476,6 +478,7 @@ mod tests {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60
         };
         let repeat = crate::repeats::RepeatInterval {
             chrom: String::from("chr7"),
@@ -507,6 +510,7 @@ mod tests {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60
         };
         let repeat = crate::repeats::RepeatInterval {
             chrom: String::from("chr7"),
@@ -537,7 +541,8 @@ mod tests {
             haploid: None,
             debug: false,
             sorted: false,
-            consensus_reads: 20
+            consensus_reads: 20,
+            max_number_reads: 60,
         };
 
         let repeat = crate::repeats::RepeatInterval {
diff --git a/src/main.rs b/src/main.rs
index 8a9c819..48f9d0d 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -82,6 +82,10 @@ pub struct Cli {
     /// Max number of reads to use to generate consensus alt sequence
     #[clap(long, value_parser, default_value_t = 20)]
     consensus_reads: usize,
+
+    // Max number of reads to extract per locus from the bam file for genotyping
+    #[clap(long, value_parser, default_value_t = 60)]
+    max_number_reads: usize,
 }
 
 fn is_file(pathname: &str) -> Result<(), String> {
diff --git a/src/parse_bam.rs b/src/parse_bam.rs
index 664d238..2a08acf 100644
--- a/src/parse_bam.rs
+++ b/src/parse_bam.rs
@@ -6,6 +6,7 @@ use rust_htslib::bam::Read;
 use std::collections::HashMap;
 use std::env;
 use url::Url;
+use rand::seq::SliceRandom;
 
 pub struct Reads {
     // could consider not to use a hashmap here and use an attribute per phase
@@ -42,6 +43,7 @@ pub fn get_overlapping_reads(
     bam: &mut bam::IndexedReader,
     repeat: &crate::repeats::RepeatInterval,
     unphased: bool,
+    max_number_reads: usize
 ) -> Option<Reads> {
     let tid = bam
         .header()
@@ -85,7 +87,26 @@ pub fn get_overlapping_reads(
         }
         None
     } else {
-        Some(Reads { seqs, ps })
+        // if more than <max_number_reads> spanning reads are found in seqs, randomly select just <max_number_reads> items from the vector
+        // if unphased, just select <max_number_reads> reads from phase 0
+        // if phased, select <max_number_reads>/2 reads from each phase
+        let mut rng = rand::thread_rng();
+        let mut seqs_filtered = HashMap::from([(0, Vec::new()), (1, Vec::new()), (2, Vec::new())]);
+        let max_reads_per_phase = HashMap::from([(0, max_number_reads), (1, max_number_reads/2), (2, max_number_reads/2)]);
+        for (phase, seqs_phase) in seqs.iter() {
+            let n_reads = seqs_phase.len();
+            let n_reads_to_select = if n_reads > max_reads_per_phase[&phase] {
+                max_reads_per_phase[&phase]
+            } else {
+                n_reads
+            };
+            let selected_reads = seqs_phase.choose_multiple(&mut rng, n_reads_to_select);
+            for read in selected_reads {
+                seqs_filtered.get_mut(phase).unwrap().push(read.to_vec());
+            }
+        }
+
+        Some(Reads { seqs: seqs_filtered, ps })
     }
 }
 
@@ -128,7 +149,7 @@ fn test_get_overlapping_reads() {
     };
     let unphased = false;
     let mut bam = create_bam_reader(&bam, &fasta);
-    let _reads = get_overlapping_reads(&mut bam, &repeat, unphased);
+    let _reads = get_overlapping_reads(&mut bam, &repeat, unphased, 60);
 }
 
 #[test]
@@ -143,7 +164,7 @@ fn test_get_overlapping_reads_url() {
     };
     let unphased = false;
     let mut bam = create_bam_reader(&bam, &fasta);
-    let _reads = get_overlapping_reads(&mut bam, &repeat, unphased);
+    let _reads = get_overlapping_reads(&mut bam, &repeat, unphased, 60);
 }
 
 #[test]
diff --git a/src/phase_insertions.rs b/src/phase_insertions.rs
index b0da167..d8cce6f 100644
--- a/src/phase_insertions.rs
+++ b/src/phase_insertions.rs
@@ -243,7 +243,7 @@ fn find_roots(
 fn find_cluster_members(
     cluster: &usize,
     cluster_to_subclusters: &HashMap<usize, (usize, usize)>,
-    insertions: &Vec<String>,
+    insertions: &[String],
 ) -> Vec<String> {
     let mut search_cluster = vec![*cluster];
     let mut insertions_of_this_cluster = vec![];
diff --git a/src/vcf.rs b/src/vcf.rs
index 92a4de3..42e204d 100644
--- a/src/vcf.rs
+++ b/src/vcf.rs
@@ -479,6 +479,7 @@ fn test_write_vcf_header_from_bam() {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60,
         };
     write_vcf_header(&args);
 }
@@ -502,6 +503,7 @@ fn test_write_vcf_header_from_name() {
             debug: false,
             sorted: false,
             consensus_reads: 20,
+            max_number_reads: 60,
         };
     write_vcf_header(&args);
 }