Add smart aligner selection and sequence length warnings

- Warns when sequences <100bp (potential alignment issues)
- Suggests SweepGA for sequences ≥100bp in verbose mode
- Improves user guidance for aligner selection
- Note: Some test files need mechanical frequency field additions (in progress)

Author: ekg

src/seqrush.rs

@@ -350,20 +350,45 @@ impl SeqRush {
             // Read alignments from PAF file
             self.align_and_unite_from_paf(paf_path, args);
         } else {
-            // Use the configured aligner backend based on CLI argument
-            match args.aligner.as_str() {
+            // Check sequence lengths and provide guidance
+            let min_len = self.sequences.iter().map(|s| s.data.len()).min().unwrap_or(0);
+            let max_len = self.sequences.iter().map(|s| s.data.len()).max().unwrap_or(0);
+
+            // Warn about very short sequences (potential alignment issues)
+            if min_len < 100 && args.verbose {
+                eprintln!("Warning: Found sequences shorter than 100 bp (min: {} bp, max: {} bp)",
+                         min_len, max_len);
+                eprintln!("         Short sequences may produce suboptimal alignments.");
+            }
+
+            // Auto-select aligner based on sequence length if using default
+            let chosen_aligner = if args.aligner == "allwave" && min_len >= 100 {
+                // Default is allwave, but for longer sequences, sweepga might be better
+                #[cfg(feature = "use-sweepga")]
+                {
+                    if args.verbose {
+                        eprintln!("Info: Sequences ≥100 bp detected. Consider --aligner sweepga for faster alignment.");
+                    }
+                }
+                "allwave"
+            } else {
+                args.aligner.as_str()
+            };
+
+            // Use the configured aligner backend
+            match chosen_aligner {
                 #[cfg(feature = "use-allwave")]
                 "allwave" => self.align_and_unite_with_allwave(args),
 
                 #[cfg(feature = "use-sweepga")]
                 "sweepga" => self.align_and_unite_with_sweepga(args),
 
                 _ => {
-                    eprintln!("Error: Unknown aligner '{}'. Available aligners:", args.aligner);
+                    eprintln!("Error: Unknown aligner '{}'. Available aligners:", chosen_aligner);
                     #[cfg(feature = "use-allwave")]
-                    eprintln!("  - allwave");
+                    eprintln!("  - allwave (works with any sequence length)");
                     #[cfg(feature = "use-sweepga")]
-                    eprintln!("  - sweepga");
+                    eprintln!("  - sweepga (requires sequences ≥20 bp, recommended for ≥100 bp)");
                     std::process::exit(1);
                 }
             }

tests/bidirected_tests.rs

@@ -72,6 +72,7 @@ fn test_simple_forward_sequence() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     run_seqrush_bidirected_simple(args).unwrap();

tests/test_complex_structural_variations.rs

@@ -111,6 +111,7 @@ fn test_inverted_repeats() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     run_inversion_aware_seqrush(args).unwrap();

tests/test_edge_traversal.rs

@@ -57,6 +57,7 @@ fn test_no_untraversed_edges_in_output() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     run_seqrush(args).unwrap();

tests/test_mathematical_correctness.rs

@@ -52,7 +52,7 @@ fn test_transitive_closure() {
         offset: 8,
     };
 
-    let mut seqrush = SeqRush::new(vec![seq1, seq2, seq3], 0);
+    let seqrush = SeqRush::new(vec![seq1, seq2, seq3], 0);
 
     // Unite seq1[0] with seq2[0]
     seqrush
@@ -90,7 +90,7 @@ fn test_single_component_per_position() {
         },
     ];
 
-    let mut seqrush = SeqRush::new(sequences, 0);
+    let seqrush = SeqRush::new(sequences, 0);
 
     // Unite matching positions
     seqrush
@@ -132,7 +132,7 @@ fn test_reverse_complement_alignment() {
         offset: 4,
     };
 
-    let mut seqrush = SeqRush::new(vec![seq1, seq2], 0);
+    let seqrush = SeqRush::new(vec![seq1, seq2], 0);
 
     // seq1 aligns to seq2 via reverse complement
     // seq1[0..4] RC aligns to seq2[0..4]
@@ -186,7 +186,7 @@ fn test_identical_sequences_produce_minimal_components() {
         },
     ];
 
-    let mut seqrush = SeqRush::new(sequences, 0);
+    let seqrush = SeqRush::new(sequences, 0);
 
     // Unite all matching positions
     seqrush
@@ -264,7 +264,7 @@ fn test_partial_alignment() {
         offset: 12,
     };
 
-    let mut seqrush = SeqRush::new(vec![seq1, seq2], 0);
+    let seqrush = SeqRush::new(vec![seq1, seq2], 0);
 
     // Only the GGGG region matches (positions 4-7 in seq1, 4-7 in seq2)
     seqrush

tests/test_rc_node_grouping.rs

@@ -45,6 +45,7 @@ fn test_rc_sequences_share_nodes() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     let output_path = args.output.clone();
@@ -136,6 +137,7 @@ fn test_node_sequence_consistency() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     let output_path = args.output.clone();

tests/test_topological_sort.rs

@@ -47,6 +47,7 @@ fn test_sorting_creates_sequential_node_ids() {
             sgd_sort: false,
             groom: false,
         aligner: "allwave".to_string(),
+        frequency: None,
     };
 
     run_seqrush(args_sorted).unwrap();
@@ -279,4 +280,4 @@ fn test_sorting_preserves_graph_structure() {
     fs::remove_file(test_fasta).ok();
     fs::remove_file(test_output_sorted).ok();
     fs::remove_file(test_output_unsorted).ok();
-}
+}