Merge branch 'DOR-427_MN_tag' into 'master'

Resolve DOR-427 "MN tag" Closes DOR-427 See merge request machine-learning/dorado!724
nanoporetech · Nov 24, 2023 · 942a35a · 942a35a
2 parents f0ac935 + 9ec47a7
commit 942a35a
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 1 deletion.
diff --git a/documentation/SAM.md b/documentation/SAM.md
@@ -42,6 +42,7 @@
 | pi:Z:  | parent read id for a split read                            |
 | sp:i:  | start coordinate of split read in parent read signal       |
 | pt:i:  | estimated poly(A/T) tail length in cDNA and dRNA reads     |
+| MN:i:  | Length of sequence at the time MM and ML were produced     |
 
 #### Modified Base Tags
 

diff --git a/dorado/read_pipeline/ReadPipeline.cpp b/dorado/read_pipeline/ReadPipeline.cpp
@@ -206,6 +206,9 @@ void ReadCommon::generate_modbase_tags(bam1_t *aln, uint8_t threshold) const {
         }
     }
 
+    int seq_len = int(seq.length());
+    bam_aux_append(aln, "MN", 'i', sizeof(seq_len), (uint8_t *)&seq_len);
+
     bam_aux_append(aln, "MM", 'Z', int(modbase_string.length() + 1),
                    (uint8_t *)modbase_string.c_str());
     bam_aux_update_array(aln, "ML", 'C', int(modbase_prob.size()), (uint8_t *)modbase_prob.data());

diff --git a/tests/test_simple_basecaller_execution.sh b/tests/test_simple_basecaller_execution.sh
@@ -52,7 +52,7 @@ if $dorado_bin basecaller ${model} $data_dir/pod5 -b ${batch} --emit-fastq --ref
     echo "Error: dorado basecaller should fail with combination of emit-fastq and reference!"
     exit 1
 fi
-if $dorado_bin basecaller ${model} $data_dir/pod5 -b ${batch} --emit-fastq --modified-bases 5mCG > $output_dir/error_condition.fq; then
+if $dorado_bin basecaller ${model} $data_dir/pod5 -b ${batch} --emit-fastq --modified-bases 5mCG_5hmCG > $output_dir/error_condition.fq; then
     echo  "Error: dorado basecaller should fail with combination of emit-fastq and modbase!"
     exit 1
 fi