diff --git a/CHANGELOG.md b/CHANGELOG.md
index f5dd1b4..0b355fb 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,9 @@
 # CHANGELOG
 
+## 1.2.0
+
+- Fix: Handle BAM records with empty SEQ fields (`*` in SAM format) in `base-depth`. Previously this would cause an out-of-bounds error. Now these reads still count toward depth, and their bases are counted as `N`. ([#92](https://github.com/sstadick/perbase/pull/92) by @ghuls)
+
 ## 1.1.0
 
 - Fix/Feat: For both MapQualBaseQualN and BaseQualMapQualN only resolve to N when the bases are ambiguous, otherwise return the consensus base.
diff --git a/Cargo.lock b/Cargo.lock
index a1e1fc3..9f1f1a2 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1027,7 +1027,7 @@ checksum = "de3145af08024dea9fa9914f381a17b8fc6034dfb00f3a84013f7ff43f29ed4c"
 
 [[package]]
 name = "perbase"
-version = "1.1.0"
+version = "1.2.0"
 dependencies = [
  "anyhow",
  "bio",
diff --git a/Cargo.toml b/Cargo.toml
index e1a0c95..bc8393b 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "perbase"
-version = "1.1.0"
+version = "1.2.0"
 authors = ["Seth Stadick <sstadick@gmail.com>"]
 edition = "2024"
 license = "MIT"
diff --git a/README.md b/README.md
index 57b8c7f..c74d762 100644
--- a/README.md
+++ b/README.md
@@ -34,6 +34,8 @@ You can also download a binary from the [releases](https://github.com/sstadick/p
 
 The `base-depth` tool walks over every position in the BAM/CRAM file and calculates the depth, as well as the number of each nucleotide at the given position. Additionally, it counts the numbers of Ins/Dels at each position.
 
+**Note on empty SEQ records:** BAM records with empty SEQ fields (represented as `*` in SAM format) are handled gracefully. These reads still count toward depth, but since the actual nucleotide cannot be determined, they are counted as `N`.
+
 The output columns are as follows:
 
 | Column         | Description                                                                                        |
diff --git a/src/lib/position/pileup_position.rs b/src/lib/position/pileup_position.rs
index d2613b1..742a7cb 100644
--- a/src/lib/position/pileup_position.rs
+++ b/src/lib/position/pileup_position.rs
@@ -107,7 +107,7 @@ impl PileupPosition {
             // Check if we are checking the base quality score
             // && Check if the base quality score is greater or equal to than the cutoff
             if let Some(base_qual_filter) = base_filter
-                && record.qual()[alignment.qpos().unwrap()] < base_qual_filter
+                && (record.seq().is_empty() || record.qual()[alignment.qpos().unwrap()] < base_qual_filter)
             {
                 self.n += 1
             } else if let Some(b) = recommended_base {
@@ -124,6 +124,8 @@ impl PileupPosition {
                     Base::M => self.m += 1,
                     _ => self.n += 1,
                 }
+            } else if record.seq().is_empty() {
+                self.n += 1
             } else {
                 match (record.seq()[alignment.qpos().unwrap()] as char).to_ascii_uppercase() {
                     'A' => self.a += 1,
@@ -517,6 +519,235 @@ mod tests {
         }
     }
 
+    /// Test that reads with empty SEQ (represented as `*` in SAM format) are handled correctly
+    /// in the pileup code path used by base-depth. Empty SEQ reads should still count toward
+    /// depth, but their bases should be counted as N. This tests the fix for issue #91.
+    #[test]
+    fn test_empty_seq_pileup() {
+        use rust_htslib::bam::{IndexedReader, Writer, index};
+        use tempfile::tempdir;
+
+        let tempdir = tempdir().unwrap();
+        let bam_path = tempdir.path().join("empty_seq.bam");
+
+        // Create header
+        let mut header = bam::header::Header::new();
+        let mut chr1 = bam::header::HeaderRecord::new(b"SQ");
+        chr1.push_tag(b"SN", &"chr1".to_owned());
+        chr1.push_tag(b"LN", &"100".to_owned());
+        header.push_record(&chr1);
+        let view = bam::HeaderView::from_header(&header);
+
+        // Create records including one with empty SEQ
+        // Normal read with 'A' bases at positions 1-25
+        let normal_record = Record::from_sam(
+            &view,
+            b"NORMAL\t0\tchr1\t1\t40\t25M\t*\t0\t0\tAAAAAAAAAAAAAAAAAAAAAAAAA\t#########################",
+        )
+        .unwrap();
+
+        // Read with empty SEQ at positions 10-35 - should count toward depth and N count
+        let empty_seq_record = Record::from_sam(
+            &view,
+            b"EMPTY_SEQ\t0\tchr1\t10\t40\t25M\t*\t0\t0\t*\t*",
+        )
+        .unwrap();
+
+        // Another normal read with 'G' bases at positions 15-40
+        let normal_record2 = Record::from_sam(
+            &view,
+            b"NORMAL2\t0\tchr1\t15\t40\t25M\t*\t0\t0\tGGGGGGGGGGGGGGGGGGGGGGGGG\t#########################",
+        )
+        .unwrap();
+
+        // Write BAM file
+        let mut writer = Writer::from_path(&bam_path, &header, bam::Format::Bam).unwrap();
+        writer.write(&normal_record).unwrap();
+        writer.write(&empty_seq_record).unwrap();
+        writer.write(&normal_record2).unwrap();
+        drop(writer);
+
+        // Index the BAM file
+        index::build(&bam_path, None, index::Type::Bai, 1).unwrap();
+
+        // Read back and create pileup
+        let mut reader = IndexedReader::from_path(&bam_path).unwrap();
+        let header_view = reader.header().clone();
+
+        let read_filter = DefaultReadFilter::new(0, 0, 0);
+
+        // Test at position 14 (0-based) = position 15 (1-based)
+        // At this position: NORMAL (A), EMPTY_SEQ (*), NORMAL2 (G) = depth 3
+        reader.fetch(("chr1", 14, 15)).unwrap();
+        let pileup_iter = reader.pileup();
+
+        for pileup_result in pileup_iter {
+            let pileup = pileup_result.unwrap();
+            if pileup.pos() == 14 {
+                let position = PileupPosition::from_pileup(pileup, &header_view, &read_filter, None);
+
+                // Depth should include all 3 reads
+                assert_eq!(position.depth, 3, "Depth should be 3 (NORMAL + EMPTY_SEQ + NORMAL2)");
+
+                // 'A' count from NORMAL read
+                assert_eq!(position.a, 1, "Should have 1 A from NORMAL read");
+
+                // 'G' count from NORMAL2 read
+                assert_eq!(position.g, 1, "Should have 1 G from NORMAL2 read");
+
+                // 'N' count from EMPTY_SEQ read (empty seq should be counted as N)
+                assert_eq!(position.n, 1, "Empty SEQ read should be counted as N");
+
+                break;
+            }
+        }
+    }
+
+    /// Test that reads with empty SEQ work correctly with base quality filtering.
+    /// When a base quality filter is set and SEQ is empty, the read should be counted as N.
+    #[test]
+    fn test_empty_seq_with_base_qual_filter() {
+        use rust_htslib::bam::{IndexedReader, Writer, index};
+        use tempfile::tempdir;
+
+        let tempdir = tempdir().unwrap();
+        let bam_path = tempdir.path().join("empty_seq_bq.bam");
+
+        // Create header
+        let mut header = bam::header::Header::new();
+        let mut chr1 = bam::header::HeaderRecord::new(b"SQ");
+        chr1.push_tag(b"SN", &"chr1".to_owned());
+        chr1.push_tag(b"LN", &"100".to_owned());
+        header.push_record(&chr1);
+        let view = bam::HeaderView::from_header(&header);
+
+        // Normal read with high quality
+        let normal_record = Record::from_sam(
+            &view,
+            b"NORMAL\t0\tchr1\t1\t40\t25M\t*\t0\t0\tAAAAAAAAAAAAAAAAAAAAAAAAA\tIIIIIIIIIIIIIIIIIIIIIIIII",
+        )
+        .unwrap();
+
+        // Read with empty SEQ
+        let empty_seq_record = Record::from_sam(
+            &view,
+            b"EMPTY_SEQ\t0\tchr1\t1\t40\t25M\t*\t0\t0\t*\t*",
+        )
+        .unwrap();
+
+        // Write BAM file
+        let mut writer = Writer::from_path(&bam_path, &header, bam::Format::Bam).unwrap();
+        writer.write(&normal_record).unwrap();
+        writer.write(&empty_seq_record).unwrap();
+        drop(writer);
+
+        // Index the BAM file
+        index::build(&bam_path, None, index::Type::Bai, 1).unwrap();
+
+        // Read back and create pileup with base quality filter
+        let mut reader = IndexedReader::from_path(&bam_path).unwrap();
+        let header_view = reader.header().clone();
+
+        let read_filter = DefaultReadFilter::new(0, 0, 0);
+
+        reader.fetch(("chr1", 0, 1)).unwrap();
+        let pileup_iter = reader.pileup();
+
+        for pileup_result in pileup_iter {
+            let pileup = pileup_result.unwrap();
+            if pileup.pos() == 0 {
+                // With base quality filter of 20
+                let position =
+                    PileupPosition::from_pileup(pileup, &header_view, &read_filter, Some(20));
+
+                // Depth should include both reads
+                assert_eq!(position.depth, 2, "Depth should be 2");
+
+                // 'A' from NORMAL (quality 'I' = 40, passes filter)
+                assert_eq!(position.a, 1, "Should have 1 A from high-quality NORMAL read");
+
+                // Empty SEQ read should be counted as N (regardless of base qual filter)
+                assert_eq!(position.n, 1, "Empty SEQ read should be counted as N");
+
+                break;
+            }
+        }
+    }
+
+    /// Test that reads with empty SEQ work correctly in mate-aware pileup mode.
+    #[test]
+    fn test_empty_seq_mate_aware() {
+        use rust_htslib::bam::{IndexedReader, Writer, index};
+        use tempfile::tempdir;
+
+        let tempdir = tempdir().unwrap();
+        let bam_path = tempdir.path().join("empty_seq_mate.bam");
+
+        // Create header
+        let mut header = bam::header::Header::new();
+        let mut chr1 = bam::header::HeaderRecord::new(b"SQ");
+        chr1.push_tag(b"SN", &"chr1".to_owned());
+        chr1.push_tag(b"LN", &"100".to_owned());
+        header.push_record(&chr1);
+        let view = bam::HeaderView::from_header(&header);
+
+        // Normal read
+        let normal_record = Record::from_sam(
+            &view,
+            b"NORMAL\t0\tchr1\t1\t40\t25M\t*\t0\t0\tAAAAAAAAAAAAAAAAAAAAAAAAA\t#########################",
+        )
+        .unwrap();
+
+        // Read with empty SEQ
+        let empty_seq_record = Record::from_sam(
+            &view,
+            b"EMPTY_SEQ\t0\tchr1\t1\t40\t25M\t*\t0\t0\t*\t*",
+        )
+        .unwrap();
+
+        // Write BAM file
+        let mut writer = Writer::from_path(&bam_path, &header, bam::Format::Bam).unwrap();
+        writer.write(&normal_record).unwrap();
+        writer.write(&empty_seq_record).unwrap();
+        drop(writer);
+
+        // Index the BAM file
+        index::build(&bam_path, None, index::Type::Bai, 1).unwrap();
+
+        // Read back and create pileup in mate-aware mode
+        let mut reader = IndexedReader::from_path(&bam_path).unwrap();
+        let header_view = reader.header().clone();
+
+        let read_filter = DefaultReadFilter::new(0, 0, 0);
+
+        reader.fetch(("chr1", 0, 1)).unwrap();
+        let pileup_iter = reader.pileup();
+
+        for pileup_result in pileup_iter {
+            let pileup = pileup_result.unwrap();
+            if pileup.pos() == 0 {
+                let position = PileupPosition::from_pileup_mate_aware(
+                    pileup,
+                    &header_view,
+                    &read_filter,
+                    None,
+                    MateResolutionStrategy::Original,
+                );
+
+                // Depth should include both reads
+                assert_eq!(position.depth, 2, "Depth should be 2");
+
+                // 'A' from NORMAL read
+                assert_eq!(position.a, 1, "Should have 1 A from NORMAL read");
+
+                // Empty SEQ read should be counted as N
+                assert_eq!(position.n, 1, "Empty SEQ read should be counted as N");
+
+                break;
+            }
+        }
+    }
+
     /// Test N strategy
     #[test]
     fn test_n_strategy() {