diff --git a/rules/cattle-flu.smk b/rules/cattle-flu.smk
index d1c4fc2..2ff3dfc 100644
--- a/rules/cattle-flu.smk
+++ b/rules/cattle-flu.smk
@@ -64,7 +64,8 @@ rule join_segments:
     # allow snakemake to choose the correct rule to run. Note that `wildcards.segment="genome"`
     # here, and for that we need alignments for 8 individual segments, which we refer to as `wildcards.genome_seg`
     input:
-        alignment = expand("results/{{subtype}}/{{segment}}/{{time}}/aligned_{genome_seg}.fasta", genome_seg=SEGMENTS) 
+        alignment = expand("results/{{subtype}}/{{segment}}/{{time}}/aligned_{genome_seg}.fasta", genome_seg=SEGMENTS),
+        metadata = metadata_by_wildcards,
     output:
         alignment = "results/{subtype}/{segment}/{time}/aligned.fasta",
         node_data = "results/{subtype}/{segment}/{time}/aligned.json",
@@ -77,7 +78,8 @@ rule join_segments:
         python scripts/join-segments.py \
             --segments {input.alignment} \
             --output {output.alignment} \
-            --output-node-data {output.node_data}
+            --output-node-data {output.node_data} \
+            --force-include $( cat {input.metadata} | csvtk filter2 -t -f '$host=="Human"' | csvtk cut -t -f strain | tail -n +2 | tr "\n" " " )
         """
 
 rule genome_metadata:
diff --git a/scripts/join-segments.py b/scripts/join-segments.py
index 2f072d9..9c54c27 100644
--- a/scripts/join-segments.py
+++ b/scripts/join-segments.py
@@ -8,6 +8,8 @@
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
     parser.add_argument('--segments', type = str, required = True, nargs='+', help = "per-segment alignments")
+    parser.add_argument('--force-include', type=str, nargs="+", required=False,
+                        help="Force include these strains regardless of how many segments have been sequenced")
     parser.add_argument('--output', type = str, required = True, help = "output whole genome alignment")
     parser.add_argument('--output-node-data', type = str, required = False, help = "output metadata in node-data JSON format")
     args = parser.parse_args()
@@ -51,8 +53,11 @@ def atgc_perc(seq):
         print("writing genome to ", args.output)
         for name,count in strain_counts.items():
             if count<7:
-                print(f"Excluding {name} as it only appears in {count} segments")
-                continue
+                if name in args.force_include:
+                    print(f"Force including {name} which would otherwise be dropped as it only appears in {count} segments")
+                else:
+                    print(f"Excluding {name} as it only appears in {count} segments")
+                    continue
             genome = "".join([sequence(seg, name) for seg in args.segments])
             node_data['nodes'][name] = {
                 "ATGC_perc": int( len([nt for nt in genome if nt in atgc])/len(genome) * 100),