Analyze Reported Results With Sources

Visual-first plan: prefix–suffix k-mer scan on a custom gene

Stepwise bioinformatics / coding / analysis plan (executable)

1. Inputs and checks (mandatory):
   • Obtain the custom gene sequence (FASTA or raw nucleotide string). Verify alphabet A/C/G/T (and optionally N). Remove non-IUPAC characters and confirm length >= k. Report orientation and provide reverse-complement option.
   • Sanity checks: sequence length, GC content, ambiguous base count. Fail early with clear messages if invalid.
2. Choose k (recommendation):
   • Interactive guidance: For a single gene (short sequence <10 kb), use k = 8–15 for nucleotide k-mers to balance uniqueness vs. occurrence; for very short genes use smaller k (4–8). Provide a quick sensitivity sweep across k values to show how unique k-mer counts change (plot k vs unique count) to pick k empirically.
3. Efficient k-mer counting (implementation options):
   • For a single sequence, an in-memory Python approach is simplest (sliding window, dictionary / collections.Counter). For many sequences or long genomes, use specialized k-mer counters (e.g., Jellyfish, KMC) — but for custom gene single-sequence scans, Python is fine and reproducible.
4. Prefix–suffix definition and algorithm:
   • Define prefix(k) := first k bases of a given window (or of sequence positions i..i+k-1), suffix(k) := last k bases (or sequence positions j..j+k-1). Here we will scan all k-mer windows and index them by their prefix and suffix strings.
   • Compute counts and locations for every k-mer. Create two maps: prefix_map: {prefix_kmer -> list of positions where it occurs as k-mer prefix}; suffix_map: {suffix_kmer -> list of positions where it occurs as k-mer suffix}.
   • To find prefix->suffix matches: for each prefix p, check if identical string exists in suffix_map (exact match), then enumerate position pairs (prefix_pos, suffix_pos) and compute distance, orientation (overlap/adjacency), and whether the match is within the same reading frame, etc.
5. Outputs to produce:
   • Table of unique k-mers with counts, prefix/suffix status, and positional list (CSV/TSV).
   • Prefix–Suffix match table listing: k-mer string, prefix positions, suffix positions, counts, nearest distance, whether overlap occurs, and whether match is cis (prefix before suffix) or trans (suffix before prefix via reverse-complement if requested).
   • Heatmap matrix (prefix set × suffix set) with counts to visualize common prefix-suffix pairs.
   • Positional track visualization: sequence coordinate plot showing where matched k-mers occur (useful for identifying repeats, palindromes, or microhomology sites).
6. Statistical checks and filters:
   • Filter low-frequency k-mers (singletons) if desired. Provide expected random-match background: compute expected frequency of a specific k-mer under i.i.d. base model (approx (1/4)^k adjusted for observed base composition) to highlight enriched k-mers.
   • Flag palindromic k-mers (k-mer == reverse_complement(k-mer)), which by definition will appear in both prefix and suffix roles and may bias counts.
7. Visualization suite (interactive):
   • k vs unique-k-mer histogram (to choose k).
   • Prefix–Suffix heatmap (counts) with interactive hover to show k-mer strings and coordinates.
   • Sequence map: linear track with colored markers for prefix occurrences and suffix occurrences and lines linking matched pairs (use Plotly scatter + shapes for links).
   • Download buttons for raw CSV/TSV, JSON of maps, and an HTML report with embedded plots.
8. Quality assurance and reproducibility:
   • Log input parameters (k, strand, filters), random seeds (if any sampling used), and versioned script with execution time & memory footprint. Provide unit tests on small example sequences (including edge cases: all-As, palindromes, ambiguous bases).

Ready-to-run Python code outline (self-contained) — run locally or via the AI Scientist agent

# Outline (not executed here):
# 1) load sequence (FASTA or raw)
# 2) sanitize & optionally reverse complement
# 3) for chosen k, slide window and record: (kmer, pos)
# 4) for each kmer record prefix=kmer[:k], suffix=kmer[-k:]; index positions in prefix_map and suffix_map
# 5) build tables and compute prefix->suffix matches
# 6) export CSV and build Plotly plots

# This outline is intentionally compact; the AI Scientist agent can run a fully-working script on your sequence when you click Run.