Add Supp. Fig. 2b JS-distance analysis: code, region/background BEDs, FASTAs, results, baseline tokenizers

.gitattributes

@@ -34,3 +34,11 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 figures/supp_fig2a_motif.png filter=lfs diff=lfs merge=lfs -text
+data/background/region_union.bed filter=lfs diff=lfs merge=lfs -text
+data/fastas/enhancer.fa filter=lfs diff=lfs merge=lfs -text
+data/fastas/exon.fa filter=lfs diff=lfs merge=lfs -text
+data/fastas/promoter.fa filter=lfs diff=lfs merge=lfs -text
+data/fastas/random.fa filter=lfs diff=lfs merge=lfs -text
+data/regions/enhancers_dels.clean.merged.bed filter=lfs diff=lfs merge=lfs -text
+figures/combined_panel_a_b.png filter=lfs diff=lfs merge=lfs -text
+figures/js_vs_random_heatmap.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,23 +1,49 @@
-# Supplementary: tokenizer motif-preservation benchmark
+# Supplementary: tokenizer characterization benchmark
 
-Code, data, and results behind Supplementary Figure 2a (perfect match rate and motif fragmentation).
+Code, data, and results behind Supplementary Figure 2:
+- **(a)** motif preservation — perfect match rate and motif fragmentation
+- **(b)** Jensen–Shannon distance between functional-region token distributions and a matched random genomic background
+
+Tokenizers compared: DNABERT-2 BPE, Grover BPE, cCRE-guided BPE, motif-guided BPE, DNAMotifTokenizer.
 
 ## Layout
-- `data/motifs.txt` — 904 unique motifs; wildcard (`*`) motifs are expanded to all A/C/G/T variants (≤5 wildcards), giving 3,737 test sequences of length 5–12.
-- `code/motif_coverage_eval.py` — tokenizes each variant (no special tokens) and computes per-tokenizer metrics.
-- `code/dna_motif_tokenizer.py` — wrapper exposing DNAMotifTokenizer through a `tokenizers`-like interface (seed = 42).
-- `code/plot_supp_fig2a_motif.py` — draws the figure from `results/summary_by_tokenizer.csv`.
-- `tokenizers/` — cCRE-guided BPE, motif-guided BPE, and DNAMotifTokenizer files. DNABERT-2 and Grover tokenizers come from their original releases.
-- `results/` — per-variant results, per-tokenizer summary, and per-motif consistency.
-- `figures/` — Supplementary Figure 2a.
-
-## Metrics
-- **Perfect match rate** — % of motif variants encoded as exactly one token.
-- **Avg. tokens per motif** — mean number of tokens per motif variant (lower = less fragmentation).
+
+### Shared
+- `tokenizers/` — all five tokenizers used in both panels
+  - `DNAbert2_tokenizer.json`, `Grover_tokenizer.json` — original DNABERT-2 / Grover tokenizers
+  - `cCRE_region_BPE_tokenizer.json`, `motif_region_BPE_tokenizer.json`
+  - `DNAMotifTokenizer/` — trie, motif list, vocab (loaded via `code/dna_motif_tokenizer.py`, seed = 42)
+- `code/dna_motif_tokenizer.py` — wrapper exposing DNAMotifTokenizer through a `tokenizers`-like interface
+
+### Panel a — motif preservation
+- `data/motifs.txt` — 904 unique motifs; wildcard (`*`) motifs are expanded to all A/C/G/T variants (≤5 wildcards), giving 3,737 test sequences of length 5–12
+- `code/motif_coverage_eval.py` — tokenizes each variant (no special tokens) and computes per-tokenizer metrics
+- `results/summary_by_tokenizer.csv` — per-tokenizer metrics plotted in the figure
+- `results/motif_variant_results.csv`, `results/consistency_by_motif.csv` — per-variant and per-motif detail
+- `code/plot_supp_fig2a_motif.py` → `figures/supp_fig2a_motif.{pdf,png}`
+
+Metrics:
+- **Perfect match rate** — % of motif variants encoded as exactly one token
+- **Avg. tokens per motif** — mean number of tokens per motif variant (lower = less fragmentation)
+
+### Panel b — region vs random-background JS distance
+- `data/regions/` — promoter (2 kb), enhancer, and exon interval sets (merged BED)
+- `data/background/` — `chrom.sizes` (canonical hg38 chromosomes), `region_union.bed` (union of the three region sets, used as exclusion), `random.bed` (exon intervals shuffled into the non-region genome with `bedtools shuffle -seed 42`, so total bp matches exon)
+- `data/fastas/` — hg38 sequences for promoter, enhancer, exon, and random (`bedtools getfasta`)
+- `code/js_vs_random_heatmap.py` — builds the background, tokenizes each FASTA, and computes the Jensen–Shannon distance (base 2) between each region's token-frequency distribution and the random background
+- `results/js_vs_random_matrix.csv` — 5 tokenizers × 3 regions matrix plotted in the figure
+- `results/js_vs_random.log` — run log with token counts per region and tokenizer
+- `figures/js_vs_random_heatmap.{pdf,png}` — heatmap from `js_vs_random_heatmap.py`
+- `code/plot_combined_panel_a_b.py` → `figures/combined_panel_a_b.{pdf,png}` — combined draft of panels a and b
 
 ## Reproduce
 ```
+# Panel a
 python code/motif_coverage_eval.py --motif-file data/motifs.txt --output-dir results
 python code/plot_supp_fig2a_motif.py
+
+# Panel b (requires bedtools and hg38.fa from UCSC)
+python code/js_vs_random_heatmap.py              # full run
+python code/js_vs_random_heatmap.py --plot-only  # redraw from results/js_vs_random_matrix.csv
 ```
-`motif_coverage_eval.py` resolves tokenizer paths relative to its original project layout (see `DEFAULT_TOKENIZER_PATHS` / `CUSTOM_TOKENIZER_PATHS`); point them at `tokenizers/` before running.
+`motif_coverage_eval.py`, `js_vs_random_heatmap.py`, and `plot_combined_panel_a_b.py` are uploaded as run and resolve paths relative to the original project layout (`DEFAULT_TOKENIZER_PATHS`, `ROOT`, `REGION_BEDS`, etc.); point them at `tokenizers/` and `data/` before running. The hg38 reference genome is not included (public, ~3 GB); the extracted FASTAs in `data/fastas/` are what the analysis tokenized.

code/js_vs_random_heatmap.py

@@ -0,0 +1,241 @@
+"""
+Simplified region-vs-random JS-distance heatmap.
+
+For each tokenizer, compute token-frequency distributions for
+promoter, enhancer, exon, and a random background (genomic shuffle
+excluding those three region sets, total bp matched to exon).
+
+Metric: Jensen-Shannon distance (sqrt of JS divergence) of each
+region distribution vs the random-background distribution.
+
+Output: 5 (tokenizers) x 3 (regions) heatmap.
+"""
+
+from __future__ import annotations
+
+import argparse
+import subprocess
+from collections import Counter
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import seaborn as sns
+from matplotlib.colors import LinearSegmentedColormap
+from scipy.spatial.distance import jensenshannon
+from tokenizers import Tokenizer
+
+ROOT = Path("/data/n5huang/dna_token")
+GENOME = ROOT / "hg38.fa"
+FAI = ROOT / "hg38.fa.fai"
+OUTDIR = ROOT / "enrichment_analysis"
+
+REGION_BEDS = {
+    "promoter": ROOT / "promoters_2kb.clean.merged.bed",
+    "enhancer": ROOT / "enhancers_dels.clean.merged.bed",
+    "exon":     ROOT / "exon.clean.merged.bed",
+}
+
+TOKENIZER_PATHS = {
+    "DNAbert2":   ROOT / "pretrain/models/DNAbert2_Pretrained/tokenizer.json",
+    "Grover":     ROOT / "pretrain/models/Grover_Pretrained/tokenizer.json",
+    "cCRE_region_BPE":  ROOT / "tokenizer_files/cCRE_region_BPE_tokenizer.json",
+    "motif_region_BPE": ROOT / "tokenizer_files/motif_region_BPE_tokenizer.json",
+}
+
+CUSTOM_TOKENIZER_PATHS = {
+    "DNAMotifTokenizer": ROOT / "tokenizer_files/DNAMotifTokenizer_official",
+}
+
+REGIONS = ["promoter", "enhancer", "exon"]
+
+TITLE_MAP = {
+    "merge_len2": "EvoLen (5120)",
+    "DNAbert2": "DNABERT-2",
+    "Grover": "Grover",
+    "cCRE_region_BPE": "cCRE-Region BPE",
+    "motif_region_BPE": "Motif-Region BPE",
+    "DNAMotifTokenizer": "DNAMotif",
+}
+
+
+def make_chrom_sizes(fai: Path, out: Path) -> None:
+    if out.exists() and out.stat().st_size > 0:
+        return
+    with fai.open() as fh, out.open("w") as oh:
+        for line in fh:
+            name, size, *_ = line.rstrip("\n").split("\t")
+            # keep only canonical autosomes + X/Y (skip alt/random/unlocalized)
+            if "_" in name or name in ("chrM", "chrEBV"):
+                continue
+            oh.write(f"{name}\t{size}\n")
+
+
+def union_bed(beds: list[Path], out: Path) -> None:
+    if out.exists() and out.stat().st_size > 0:
+        return
+    cmd = "cat " + " ".join(str(b) for b in beds) + \
+          f" | cut -f1-3 | bedtools sort -i - | bedtools merge -i - > {out}"
+    subprocess.run(cmd, shell=True, check=True)
+
+
+def shuffle_to_random(exon_bed: Path, excl_bed: Path, chrom_sizes: Path, out: Path) -> None:
+    """Shuffle exon intervals into genome excluding region union -> same bp as exon."""
+    if out.exists() and out.stat().st_size > 0:
+        return
+    subprocess.run(
+        f"bedtools shuffle -i {exon_bed} -g {chrom_sizes} -excl {excl_bed} -seed 42 "
+        f"| bedtools sort -i - | bedtools merge -i - > {out}",
+        shell=True, check=True,
+    )
+
+
+def get_fasta(bed: Path, out_fa: Path) -> None:
+    if out_fa.exists() and out_fa.stat().st_size > 0:
+        return
+    subprocess.run(
+        f"bedtools getfasta -fi {GENOME} -bed {bed} -fo {out_fa}",
+        shell=True, check=True,
+    )
+
+
+def tokenize_fasta(fasta: Path, tok: Tokenizer, batch_size: int = 2000) -> Counter:
+    counts: Counter = Counter()
+    sequences: list[str] = []
+    seq_parts: list[str] = []
+    with fasta.open() as f:
+        for line in f:
+            line = line.strip()
+            if line.startswith(">"):
+                if seq_parts:
+                    sequences.append("".join(seq_parts))
+                    seq_parts = []
+            else:
+                seq_parts.append(line.upper().replace("N", ""))
+    if seq_parts:
+        sequences.append("".join(seq_parts))
+    sequences = [s for s in sequences if s]
+
+    for i in range(0, len(sequences), batch_size):
+        batch = sequences[i:i + batch_size]
+        for enc in tok.encode_batch(batch):
+            counts.update(enc.tokens)
+    return counts
+
+
+def js_distance(p_counts: Counter, q_counts: Counter) -> float:
+    vocab = sorted(set(p_counts) | set(q_counts))
+    p = np.array([p_counts.get(t, 0) for t in vocab], dtype=float)
+    q = np.array([q_counts.get(t, 0) for t in vocab], dtype=float)
+    p_sum = p.sum() or 1.0
+    q_sum = q.sum() or 1.0
+    p /= p_sum
+    q /= q_sum
+    # scipy's jensenshannon already returns the distance (sqrt of divergence), base=2 below
+    return float(jensenshannon(p, q, base=2))
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--plot-only", action="store_true")
+    args = parser.parse_args()
+
+    OUTDIR.mkdir(exist_ok=True)
+    beds_dir = OUTDIR / "beds_simple"
+    fasta_dir = OUTDIR / "fastas_simple"
+    beds_dir.mkdir(exist_ok=True)
+    fasta_dir.mkdir(exist_ok=True)
+
+    matrix_csv = OUTDIR / "js_vs_random_matrix.csv"
+
+    if args.plot_only:
+        df = pd.read_csv(matrix_csv, index_col=0)
+    else:
+        chrom_sizes = beds_dir / "chrom.sizes"
+        make_chrom_sizes(FAI, chrom_sizes)
+
+        # Build region union to use as exclusion
+        excl = beds_dir / "region_union.bed"
+        union_bed(list(REGION_BEDS.values()), excl)
+
+        # Shuffle exon into non-region genome -> "random" background
+        random_bed = beds_dir / "random.bed"
+        shuffle_to_random(REGION_BEDS["exon"], excl, chrom_sizes, random_bed)
+
+        # Extract FASTA for each set
+        fastas: dict[str, Path] = {}
+        for name, bed in REGION_BEDS.items():
+            fa = fasta_dir / f"{name}.fa"
+            get_fasta(bed, fa)
+            fastas[name] = fa
+        fa_rand = fasta_dir / "random.fa"
+        get_fasta(random_bed, fa_rand)
+        fastas["random"] = fa_rand
+
+        rows = []
+        tok_names_ordered = list(TOKENIZER_PATHS.keys()) + list(CUSTOM_TOKENIZER_PATHS.keys())
+        # Load HF tokenizers
+        loaded = {n: Tokenizer.from_file(str(p)) for n, p in TOKENIZER_PATHS.items()}
+        # Load custom tokenizers
+        import sys
+        sys.path.insert(0, str(ROOT / "tokenizer_evaluation"))
+        from dna_motif_tokenizer import DNAMotifTokenizer, Trie, TrieNode
+        sys.modules["__main__"].Trie = Trie
+        sys.modules["__main__"].TrieNode = TrieNode
+        for n, p in CUSTOM_TOKENIZER_PATHS.items():
+            loaded[n] = DNAMotifTokenizer(str(p), seed=42)
+
+        for tok_name in tok_names_ordered:
+            print(f"\nTokenizing with {tok_name}...")
+            tok = loaded[tok_name]
+            counts = {}
+            for name, fa in fastas.items():
+                counts[name] = tokenize_fasta(fa, tok)
+                print(f"  {name}: {sum(counts[name].values()):,} tokens")
+            row = {r: js_distance(counts[r], counts["random"]) for r in REGIONS}
+            rows.append(row)
+        df = pd.DataFrame(rows, index=tok_names_ordered, columns=REGIONS)
+        df.to_csv(matrix_csv)
+        print(f"\nSaved matrix: {matrix_csv}")
+        print(df.round(4))
+
+    # Plot
+    display_index = [TITLE_MAP.get(n, n) for n in df.index]
+    df_display = df.copy()
+    df_display.index = display_index
+
+    cmap = LinearSegmentedColormap.from_list("evo_heat", ["white", "#7EDCC5", "#636E72"])
+    fig, ax = plt.subplots(figsize=(22, 32), constrained_layout=True)
+    sns.heatmap(
+        df_display,
+        ax=ax,
+        annot=True,
+        fmt=".4f",
+        annot_kws={"size": 56, "fontweight": "bold"},
+        cmap=cmap,
+        vmin=0.0,
+        vmax=float(df_display.values.max()) or 0.001,
+        linewidths=0.5,
+        cbar_kws={"label": "Jensen-Shannon distance"},
+    )
+    ax.set_title("Region vs random-genome token-frequency separation",
+                 fontsize=60, fontweight="bold", pad=30)
+    ax.set_xlabel("Genomic region", fontsize=56)
+    ax.set_ylabel("Tokenizer", fontsize=56)
+    ax.tick_params(axis="x", labelsize=48)
+    ax.tick_params(axis="y", labelsize=48, rotation=0)
+    cbar = ax.collections[0].colorbar
+    cbar.ax.tick_params(labelsize=48)
+    cbar.set_label("Jensen-Shannon distance", size=56)
+
+    out_pdf = OUTDIR / "js_vs_random_heatmap.pdf"
+    out_png = OUTDIR / "js_vs_random_heatmap.png"
+    fig.savefig(out_pdf, dpi=300, bbox_inches="tight")
+    fig.savefig(out_png, dpi=300, bbox_inches="tight")
+    print(f"Saved: {out_pdf}")
+    print(f"Saved: {out_png}")
+
+
+if __name__ == "__main__":
+    main()

code/plot_combined_panel_a_b.py

@@ -0,0 +1,154 @@
+#!/usr/bin/env python3
+"""Combined figure: (a) Motif Preservation (A1 + A2 stacked) and (b) JS heatmap, in one row."""
+
+import os
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import matplotlib as mpl
+import seaborn as sns
+from matplotlib.colors import LinearSegmentedColormap
+from matplotlib.gridspec import GridSpec, GridSpecFromSubplotSpec
+
+# ── Data ──
+df_summary = pd.read_csv(
+    "/data/n5huang/dna_token/tokenizer_evaluation/motif_eval_outputs_new/summary_by_tokenizer.csv"
+)
+df_js = pd.read_csv(
+    "/data/n5huang/dna_token/enrichment_analysis/js_vs_random_matrix.csv",
+    index_col=0,
+)
+
+# ── Method config (shared) ──
+METHODS = ["DNAbert2", "Grover", "cCRE_region_BPE", "motif_region_BPE", "DNAMotifTokenizer"]
+DISPLAY_NAMES = {
+    "DNAbert2": "DNABERT-2",
+    "Grover": "Grover",
+    "cCRE_region_BPE": "cCRE-Region BPE",
+    "motif_region_BPE": "Motif-Region BPE",
+    "DNAMotifTokenizer": "DNAMotif",
+}
+METHOD_COLORS = {
+    "DNAbert2": "#6C5CE7",          # purple (from C)
+    "Grover": "#E74C3C",            # red (from T)
+    "cCRE_region_BPE": "#52BE80",   # green (from A)
+    "motif_region_BPE": "#F4D03F",  # yellow (Token Characterization)
+    "DNAMotifTokenizer": "#E67E22", # orange (Fine-tuning Ability)
+}
+HEATMAP_REGIONS = ["promoter", "enhancer", "exon"]
+JS_INDEX_DISPLAY = [DISPLAY_NAMES.get(n, n) for n in df_js.index]
+
+# ── Plot style ──
+mpl.rcParams.update({
+    "font.family": "sans-serif",
+    "font.size": 58,
+    "axes.labelsize": 64,
+    "axes.titlesize": 66,
+    "axes.titleweight": "bold",
+    "xtick.labelsize": 58,
+    "ytick.labelsize": 58,
+})
+
+
+def get_value(method, source_df, col_name):
+    row = source_df[source_df["Tokenizer"] == method]
+    if len(row) == 0:
+        return float("nan")
+    return row[col_name].values[0]
+
+
+def draw_bar_panel(ax, title, ylabel, col_name, higher_better, title_pad=30, ylabel_y=0.5):
+    vals = [get_value(m, df_summary, col_name) for m in METHODS]
+    colors = [METHOD_COLORS[m] for m in METHODS]
+    x_labels = [DISPLAY_NAMES[m] for m in METHODS]
+    x = np.arange(len(METHODS)) * 0.3
+    bars = ax.bar(x, vals, color=colors, edgecolor="white", linewidth=0.5, alpha=0.85, width=0.2)
+
+    for i, v in enumerate(vals):
+        if np.isnan(v):
+            continue
+        y_off = max(vals) * 0.02
+        fmt = f"{v:.2f}" if abs(v) >= 1 else f"{v:.4f}"
+        ax.text(x[i], v + y_off, fmt, ha="center", va="bottom", fontsize=62, fontweight="bold")
+
+    ax.set_xticks(x)
+    ax.set_xticklabels(x_labels, rotation=15, ha="right")
+    ax.set_xlim(x[0] - 0.2, x[-1] + 0.2)
+    ax.set_ylabel(ylabel)
+    # Pin ylabel position in axes coords so A1 and A2 line up regardless of tick-label widths.
+    ax.yaxis.set_label_coords(-0.08, ylabel_y)
+    ax.set_title(title, pad=title_pad)
+    ax.tick_params(axis="x", pad=10)
+    ax.tick_params(axis="y", pad=15)
+    ax.spines["top"].set_visible(False)
+    ax.spines["right"].set_visible(False)
+    ax.grid(axis="y", alpha=0.3, linewidth=0.8)
+
+    valid = [v for v in vals if not np.isnan(v)]
+    if valid:
+        best = (np.nanargmax if higher_better else np.nanargmin)(vals)
+        bars[best].set_edgecolor("black")
+        bars[best].set_linewidth(2)
+
+
+# ── Figure layout: 1 row, 2 top-level columns ──
+fig = plt.figure(figsize=(50, 26))
+gs = GridSpec(1, 2, figure=fig, width_ratios=[1, 0.9], wspace=0.30)
+
+# Left column (panel a): 2 stacked subplots
+gs_a = GridSpecFromSubplotSpec(2, 1, subplot_spec=gs[0, 0], hspace=0.8)
+ax_a1 = fig.add_subplot(gs_a[0, 0])
+ax_a2 = fig.add_subplot(gs_a[1, 0])
+
+# Right column (panel b): heatmap
+ax_b = fig.add_subplot(gs[0, 1])
+
+# ── Draw panel a (A1 + A2) ──
+draw_bar_panel(ax_a1, "Perfect Match Rate", "Perfect match rate (%)",
+               "Perfect_Match_Rate", higher_better=True, title_pad=80, ylabel_y=0.35)
+draw_bar_panel(ax_a2, "Motif Fragmentation", "Avg tokens per motif",
+               "Avg_Tokens_Per_Motif", higher_better=False, title_pad=80)
+
+# ── Draw panel b (heatmap) ──
+df_disp = df_js.copy()
+df_disp.index = JS_INDEX_DISPLAY
+cmap = LinearSegmentedColormap.from_list("evo_heat", ["white", "#AED6F1", "#5DADE2", "#2E86C1"])
+sns.heatmap(
+    df_disp,
+    ax=ax_b,
+    annot=True,
+    fmt=".4f",
+    annot_kws={"size": 62, "fontweight": "bold"},  # matches bar-value annotation size
+    cmap=cmap,
+    vmin=0.0,
+    vmax=float(df_disp.values.max()) or 0.001,
+    linewidths=0.5,
+    cbar_kws={"label": "Jensen-Shannon distance"},
+)
+ax_b.set_title("Region vs random-genome separation",
+               fontsize=66, fontweight="bold", pad=80)
+ax_b.set_xlabel("Genomic region", fontsize=64, labelpad=15)
+ax_b.set_ylabel("")
+ax_b.tick_params(axis="x", labelsize=58)
+ax_b.tick_params(axis="y", labelsize=58, rotation=0)
+cbar = ax_b.collections[0].colorbar
+cbar.ax.tick_params(labelsize=58)
+cbar.set_label("Jensen-Shannon distance", size=64)
+
+fig.subplots_adjust(left=0.06, right=0.98, top=0.88, bottom=0.10)
+
+# ─��� Panel labels (a), (b) on same line as titles ──
+# Use axes-relative y so they line up with titles (which sit just above each axes top).
+ax_a1.text(-0.08, 1.234, "(a)", transform=ax_a1.transAxes,
+           fontsize=66, fontweight="bold", va="center", ha="center")
+ax_b.text(-0.30, 1.083, "(b)", transform=ax_b.transAxes,
+          fontsize=66, fontweight="bold", va="center", ha="center")
+
+out_dir = "/data/n5huang/dna_token/paper_figures"
+os.makedirs(out_dir, exist_ok=True)
+out = os.path.join(out_dir, "combined_panel_a_b.pdf")
+fig.savefig(out, dpi=300)
+fig.savefig(out.replace(".pdf", ".png"), dpi=80)
+print(f"Saved: {out}")
+print(f"Saved: {out.replace('.pdf', '.png')}")
+plt.close(fig)

data/background/chrom.sizes

@@ -0,0 +1,592 @@
+chr1	248956422
+chr2	242193529
+chr3	198295559
+chr4	190214555
+chr5	181538259
+chr6	170805979
+chr7	159345973
+chr8	145138636
+chr9	138394717
+chr10	133797422
+chr11	135086622
+chr12	133275309
+chr13	114364328
+chr14	107043718
+chr15	101991189
+chr16	90338345
+chr17	83257441
+chr18	80373285
+chr19	58617616
+chr20	64444167
+chr21	46709983
+chr22	50818468
+chrX	156040895
+chrY	57227415
+GL000008.2	209709
+GL000009.2	201709
+GL000194.1	191469
+GL000195.1	182896
+GL000205.2	185591
+GL000208.1	92689
+GL000213.1	164239
+GL000214.1	137718
+GL000216.2	176608
+GL000218.1	161147
+GL000219.1	179198
+GL000220.1	161802
+GL000221.1	155397
+GL000224.1	179693
+GL000225.1	211173
+GL000226.1	15008
+KQ759759.1	196940
+KN538364.1	415308
+KQ759762.1	101037
+KQ031383.1	467143
+KN538369.1	541038
+JH159136.1	200998
+JH159137.1	191409
+KZ208923.1	48370
+KZ208924.1	209722
+KQ031387.1	320750
+KV766195.1	140877
+KZ208916.1	1046838
+KN538360.1	460100
+KZ208920.1	690932
+KZ208906.1	330031
+KN196484.1	370917
+KN196476.1	305979
+KQ983257.1	230434
+KN196479.1	330164
+KV575245.1	154723
+KZ208917.1	64689
+KZ208911.1	242796
+KN196473.1	166200
+KZ559108.1	305244
+KN196487.1	101150
+KQ759760.1	315610
+KN196475.1	451168
+KV880766.1	156998
+KV880767.1	265876
+KQ090016.1	245716
+KV880764.1	142129
+KN538361.1	305542
+KN196474.1	122022
+KZ559109.1	279644
+KQ090022.1	181958
+KN196478.1	268330
+KZ559104.1	105527
+KN196480.1	277797
+KQ090028.1	407387
+KN196483.1	35455
+KN196481.1	108875
+KN538363.1	365499
+KN538362.1	208149
+KQ031385.1	373699
+KV766192.1	411654
+KQ031386.1	165718
+KQ031388.1	179932
+KN538365.1	14347
+KN538366.1	85284
+KN538367.1	420164
+KN538370.1	86533
+KN538373.1	148762
+KZ559113.1	480415
+KV880765.1	468267
+KV766196.1	281919
+KN538371.1	206320
+KQ031384.1	481245
+KN538372.1	356766
+KQ090021.1	264545
+KN196482.1	211377
+KZ559115.1	230843
+KZ208914.1	165120
+KZ208922.1	93070
+KQ458386.1	405389
+KV575244.1	673059
+KZ559100.1	44955
+KZ208912.1	589656
+KZ208915.1	6367528
+KV880768.1	1927115
+KN196472.1	186494
+GL383545.1	179254
+GL383546.1	309802
+KI270824.1	181496
+KI270825.1	188315
+KQ090020.1	185507
+GL383547.1	154407
+KN538368.1	203552
+KI270826.1	186169
+KI270827.1	67707
+KZ559111.1	181167
+KI270829.1	204059
+KI270830.1	177092
+KI270831.1	296895
+KI270832.1	210133
+KI270902.1	106711
+KI270903.1	214625
+KZ559110.1	301637
+KI270927.1	218612
+GL877875.1	167313
+GL383549.1	120804
+GL383550.2	169178
+KQ090023.1	109323
+GL877876.1	408271
+GL383552.1	138655
+KI270904.1	572349
+GL383553.2	152874
+KI270835.1	238139
+GL383551.1	184319
+KI270837.1	40090
+KI270833.1	76061
+KI270834.1	119498
+KI270836.1	56134
+KZ208918.1	174808
+KZ559112.1	154139
+KI270838.1	306913
+KI270839.1	180306
+KI270840.1	191684
+KI270841.1	169134
+KI270842.1	37287
+KI270843.1	103832
+KQ090024.1	168146
+KQ090025.1	123480
+KI270844.1	322166
+KI270845.1	180703
+KI270846.1	1351393
+KI270847.1	1511111
+KZ208919.1	171798
+KI270852.1	478999
+KI270848.1	327382
+GL383554.1	296527
+KI270906.1	196384
+GL383555.2	388773
+KI270851.1	263054
+KI270849.1	244917
+KI270905.1	5161414
+KI270850.1	430880
+KQ031389.1	2365364
+KI270853.1	2659700
+GL383556.1	192462
+GL383557.1	89672
+KI270855.1	232857
+KQ031390.1	169136
+KI270856.1	63982
+KQ090027.1	267463
+KQ090026.1	59016
+KZ208921.1	78609
+KI270854.1	134193
+KI270909.1	325800
+KV766197.1	246895
+KZ559114.1	116753
+GL383563.3	375691
+KI270861.1	196688
+GL383564.2	133151
+GL000258.2	1821992
+KI270860.1	178921
+KI270907.1	137721
+KI270862.1	391357
+GL383565.1	223995
+KI270908.1	1423190
+KV766198.1	276292
+KI270910.1	157099
+GL383566.1	90219
+JH159146.1	278131
+JH159147.1	70345
+JH159148.1	88070
+KI270857.1	2877074
+KI270858.1	235827
+KI270859.1	108763
+KZ559116.1	163186
+GL383567.1	289831
+GL383568.1	104552
+GL383569.1	167950
+GL383570.1	164789
+GL383571.1	198278
+GL383572.1	159547
+KI270863.1	167999
+KI270864.1	111737
+KQ458385.1	205101
+KI270912.1	174061
+KI270911.1	157710
+KV575254.1	99845
+KV575246.1	163926
+KV575256.1	223118
+KV575253.1	166713
+KV575252.1	178197
+KV575255.1	161095
+KV575259.1	171263
+KI270917.1	190932
+KI270918.1	123111
+KI270919.1	170701
+KV575247.1	170206
+KV575248.1	168131
+KV575250.1	241058
+KV575249.1	293522
+KV575257.1	100553
+KI270920.1	198005
+KI270921.1	282224
+KI270922.1	187935
+KI270923.1	189352
+KI270929.1	186203
+KI270930.1	200773
+KI270931.1	170148
+KI270932.1	215732
+KI270933.1	170537
+KI270882.1	248807
+KI270883.1	170399
+KI270884.1	157053
+KI270885.1	171027
+KI270886.1	204239
+KI270887.1	209512
+KI270888.1	155532
+KV575258.1	156965
+KV575251.1	159285
+KV575260.1	145691
+KI270889.1	170698
+KI270890.1	184499
+GL000209.2	177381
+KI270891.1	170680
+KI270914.1	205194
+KI270915.1	170665
+KI270916.1	184516
+GL949746.1	987716
+GL949747.2	729520
+GL949748.2	1064304
+GL949749.2	1091841
+GL949750.2	1066390
+GL949751.2	1002683
+GL949752.1	987100
+GL949753.2	796479
+GL383573.1	385657
+GL383574.1	155864
+GL383575.2	170222
+KI270866.1	43156
+GL383576.1	188024
+KI270867.1	233762
+KI270865.1	52969
+KI270938.1	1066800
+KI270868.1	61734
+KI270760.1	109528
+KI270762.1	354444
+GL383518.1	182439
+KI270759.1	425601
+KI270766.1	256271
+GL383519.1	110268
+KI270761.1	165834
+KQ458382.1	141019
+GL383520.2	366580
+KI270763.1	911658
+KQ458383.1	349938
+KI270765.1	185285
+KI270764.1	50258
+KQ983255.1	278659
+KQ458384.1	212205
+KV880763.1	551020
+KZ208904.1	166136
+KZ208905.1	140355
+KI270892.1	162212
+GL383577.2	128386
+KI270869.1	118774
+KI270870.1	183433
+KI270871.1	58661
+GL383578.2	63917
+GL383579.2	201197
+GL383580.2	74653
+GL383581.2	116689
+KI270872.1	82692
+KI270873.1	143900
+KI270874.1	166743
+GL383582.2	162811
+GL383583.2	96924
+KI270875.1	259914
+KI270876.1	263666
+KI270877.1	101331
+KI270878.1	186262
+KI270879.1	304135
+KB663609.1	74013
+KI270928.1	176103
+KN196485.1	156562
+KN196486.1	153027
+KQ458387.1	155930
+KQ458388.1	174749
+KQ759761.1	145162
+KI270769.1	120616
+KI270767.1	161578
+GL383521.1	143390
+KI270772.1	133041
+GL383522.1	123821
+KI270770.1	136240
+KI270893.1	161218
+KI270894.1	214158
+GL582966.2	96131
+KI270773.1	70887
+KI270776.1	174166
+KI270768.1	110099
+KI270774.1	223625
+KI270771.1	110395
+KI270775.1	138019
+KQ983256.1	535088
+KZ208907.1	181658
+KZ208908.1	140361
+JH636055.2	173151
+GL383526.1	180671
+KI270779.1	205312
+KI270777.1	173649
+KI270782.1	162429
+KI270783.1	109187
+KI270778.1	248252
+KI270895.1	162896
+KZ208909.1	175849
+KI270780.1	224108
+KI270924.1	166540
+KI270781.1	113034
+KI270934.1	163458
+KZ559105.1	195063
+KI270935.1	197351
+KZ559101.1	164041
+KI270936.1	164170
+KZ559102.1	197752
+KI270937.1	165607
+KZ559103.1	302885
+KI270784.1	184404
+KQ983258.1	205407
+KV766193.1	420675
+GL383527.1	164536
+KI270790.1	220246
+GL383528.1	376187
+KI270787.1	111943
+GL000257.2	586476
+KI270785.1	119912
+KQ090013.1	90922
+KI270786.1	244096
+KI270788.1	158965
+KI270789.1	205944
+KI270896.1	378547
+KI270925.1	555799
+KQ090014.1	163749
+KQ090015.1	236512
+GL383532.1	82728
+KI270897.1	1144418
+GL383531.1	173459
+GL949742.1	226852
+GL339449.2	1612928
+KI270795.1	131892
+KI270791.1	195710
+GL383530.1	101241
+KI270898.1	130957
+KI270792.1	179043
+KI270796.1	172708
+KI270793.1	126136
+KI270794.1	164558
+KN196477.1	139087
+KV575243.1	362221
+KZ208910.1	135987
+KQ090017.1	82315
+GL383533.1	124736
+KB021644.2	185823
+KI270797.1	197536
+KI270798.1	271782
+KI270799.1	152148
+KI270800.1	175808
+KI270801.1	870480
+KI270802.1	75005
+KI270758.1	76752
+GL000250.2	4672374
+GL000251.2	4795265
+GL000252.2	4604811
+GL000253.2	4677643
+GL000254.2	4827813
+GL000255.2	4606388
+GL000256.2	4929269
+KI270804.1	157952
+KI270806.1	158166
+GL383534.2	119183
+KI270805.1	209988
+KI270899.1	190869
+KI270809.1	209586
+KI270803.1	1111570
+KI270807.1	126434
+KZ559106.1	172555
+KZ208913.1	680662
+KI270808.1	271455
+KI270811.1	292436
+KI270814.1	141812
+KI270810.1	374415
+KI270812.1	282736
+KI270815.1	132244
+KI270813.1	300230
+KI270816.1	305841
+KI270818.1	145606
+KI270817.1	158983
+KI270900.1	318687
+KI270819.1	133535
+KI270901.1	136959
+KI270820.1	36640
+KI270926.1	229282
+KZ559107.1	103072
+KI270821.1	985506
+KI270822.1	624492
+GL383539.1	162988
+GL383540.1	71551
+GL383541.1	171286
+GL383542.1	60032
+KI270823.1	439082
+KQ090018.1	163882
+KQ090019.1	134099
+KI270880.1	284869
+KI270881.1	144206
+KI270913.1	274009
+KV766199.1	188004
+KI270302.1	2274
+KI270303.1	1942
+KI270304.1	2165
+KI270305.1	1472
+KI270310.1	1201
+KI270311.1	12399
+KI270312.1	998
+KI270315.1	2276
+KI270316.1	1444
+KI270317.1	37690
+KI270320.1	4416
+KI270322.1	21476
+KI270329.1	1040
+KI270330.1	1652
+KI270333.1	2699
+KI270334.1	1368
+KI270335.1	1048
+KI270336.1	1026
+KI270337.1	1121
+KI270338.1	1428
+KI270340.1	1428
+KI270362.1	3530
+KI270363.1	1803
+KI270364.1	2855
+KI270366.1	8320
+KI270371.1	2805
+KI270372.1	1650
+KI270373.1	1451
+KI270374.1	2656
+KI270375.1	2378
+KI270376.1	1136
+KI270378.1	1048
+KI270379.1	1045
+KI270381.1	1930
+KI270382.1	4215
+KI270383.1	1750
+KI270384.1	1658
+KI270385.1	990
+KI270386.1	1788
+KI270387.1	1537
+KI270388.1	1216
+KI270389.1	1298
+KI270390.1	2387
+KI270391.1	1484
+KI270392.1	971
+KI270393.1	1308
+KI270394.1	970
+KI270395.1	1143
+KI270396.1	1880
+KI270411.1	2646
+KI270412.1	1179
+KI270414.1	2489
+KI270417.1	2043
+KI270418.1	2145
+KI270419.1	1029
+KI270420.1	2321
+KI270422.1	1445
+KI270423.1	981
+KI270424.1	2140
+KI270425.1	1884
+KI270429.1	1361
+KI270435.1	92983
+KI270438.1	112505
+KI270442.1	392061
+KI270448.1	7992
+KI270465.1	1774
+KI270466.1	1233
+KI270467.1	3920
+KI270468.1	4055
+KI270507.1	5353
+KI270508.1	1951
+KI270509.1	2318
+KI270510.1	2415
+KI270511.1	8127
+KI270512.1	22689
+KI270515.1	6361
+KI270516.1	1300
+KI270517.1	3253
+KI270518.1	2186
+KI270519.1	138126
+KI270521.1	7642
+KI270522.1	5674
+KI270528.1	2983
+KI270529.1	1899
+KI270530.1	2168
+KI270538.1	91309
+KI270539.1	993
+KI270544.1	1202
+KI270548.1	1599
+KI270579.1	31033
+KI270580.1	1553
+KI270581.1	7046
+KI270582.1	6504
+KI270583.1	1400
+KI270584.1	4513
+KI270587.1	2969
+KI270588.1	6158
+KI270589.1	44474
+KI270590.1	4685
+KI270591.1	5796
+KI270593.1	3041
+KI270706.1	175055
+KI270707.1	32032
+KI270708.1	127682
+KI270709.1	66860
+KI270710.1	40176
+KI270711.1	42210
+KI270712.1	176043
+KI270713.1	40745
+KI270714.1	41717
+KI270715.1	161471
+KI270716.1	153799
+KI270717.1	40062
+KI270718.1	38054
+KI270719.1	176845
+KI270720.1	39050
+KI270721.1	100316
+KI270722.1	194050
+KI270723.1	38115
+KI270724.1	39555
+KI270725.1	172810
+KI270726.1	43739
+KI270727.1	448248
+KI270728.1	1872759
+KI270729.1	280839
+KI270730.1	112551
+KI270731.1	150754
+KI270732.1	41543
+KI270733.1	179772
+KI270734.1	165050
+KI270735.1	42811
+KI270736.1	181920
+KI270737.1	103838
+KI270738.1	99375
+KI270739.1	73985
+KI270740.1	37240
+KI270741.1	157432
+KI270742.1	186739
+KI270743.1	210658
+KI270744.1	168472
+KI270745.1	41891
+KI270746.1	66486
+KI270747.1	198735
+KI270748.1	93321
+KI270749.1	158759
+KI270750.1	148850
+KI270751.1	150742
+KI270752.1	27745
+KI270753.1	62944
+KI270754.1	40191
+KI270755.1	36723
+KI270756.1	79590
+KI270757.1	71251

data/background/region_union.bed

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:170c641b152b0405def079ef1a79cfdca6a6fc915f449cab7cecc6a8dbd4e84c
+size 39227263

data/fastas/enhancer.fa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f98c1d46ae623caf1f128fe47da58746fa3cb2587d24af058cac8dea94d2ed7
+size 432577659

data/fastas/exon.fa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f7a41eaac27fbcefc0ed229e19c1e7fb97d7f5311858841bd45e2ef9f94c14
+size 202189721

data/fastas/promoter.fa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f9d5c0997ee9662710fd0ba41e00db70a82bde794a33e5ae0f439e3a887b4be
+size 61863838

data/fastas/random.fa

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3624642aa274d4162f0d2defb9b9d91d28e292c7a5b518448b34e37c7720beba
+size 192538591

data/regions/enhancers_dels.clean.merged.bed

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05d56b3f533818c9b8e283c107b2f7aa7176d4140625f1cf97b04e9915ead343
+size 35011481

results/js_vs_random.log

@@ -0,0 +1,40 @@
+
+Tokenizing with DNAbert2...
+  promoter: 12,334,449 tokens
+  enhancer: 81,613,472 tokens
+  exon: 39,014,803 tokens
+  random: 32,378,182 tokens
+
+Tokenizing with Grover...
+  promoter: 15,845,529 tokens
+  enhancer: 102,460,422 tokens
+  exon: 49,411,407 tokens
+  random: 41,049,736 tokens
+
+Tokenizing with cCRE_region_BPE...
+  promoter: 12,016,320 tokens
+  enhancer: 77,428,659 tokens
+  exon: 37,654,355 tokens
+  random: 31,601,798 tokens
+
+Tokenizing with motif_region_BPE...
+  promoter: 13,091,114 tokens
+  enhancer: 83,353,540 tokens
+  exon: 40,627,217 tokens
+  random: 33,666,692 tokens
+
+Tokenizing with DNAMotifTokenizer...
+  promoter: 19,543,735 tokens
+  enhancer: 126,777,588 tokens
+  exon: 61,613,820 tokens
+  random: 52,419,078 tokens
+
+Saved matrix: /data/n5huang/dna_token/enrichment_analysis/js_vs_random_matrix.csv
+                   promoter  enhancer    exon
+DNAbert2             0.2700    0.1530  0.1658
+Grover               0.2465    0.1292  0.1440
+cCRE_region_BPE      0.2677    0.1527  0.1705
+motif_region_BPE     0.2575    0.1444  0.1626
+DNAMotifTokenizer    0.2047    0.1049  0.1227
+Saved: /data/n5huang/dna_token/enrichment_analysis/js_vs_random_heatmap.pdf
+Saved: /data/n5huang/dna_token/enrichment_analysis/js_vs_random_heatmap.png

results/js_vs_random_matrix.csv

@@ -0,0 +1,6 @@
+,promoter,enhancer,exon
+DNAbert2,0.2700467663332958,0.1530054987812049,0.16576869998910815
+Grover,0.24645834442494668,0.12922598364042234,0.14397131314322673
+cCRE_region_BPE,0.2676769058033193,0.1526687523936217,0.17054039887560096
+motif_region_BPE,0.2574520985639099,0.1443754257311266,0.16263857681147068
+DNAMotifTokenizer,0.20469228698318975,0.10486649424832019,0.12271872972266289

tokenizers/Grover_tokenizer.json

@@ -0,0 +1,1359 @@
+{
+  "version": "1.0",
+  "truncation": null,
+  "padding": null,
+  "added_tokens": [
+    {
+      "id": 0,
+      "content": "[PAD]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 1,
+      "content": "[UNK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 2,
+      "content": "[CLS]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 3,
+      "content": "[SEP]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 4,
+      "content": "[MASK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  ],
+  "normalizer": null,
+  "pre_tokenizer": {
+    "type": "Whitespace"
+  },
+  "post_processor": {
+    "type": "TemplateProcessing",
+    "single": [
+      {
+        "SpecialToken": {
+          "id": "[CLS]",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "SpecialToken": {
+          "id": "[SEP]",
+          "type_id": 0
+        }
+      }
+    ],
+    "pair": [
+      {
+        "SpecialToken": {
+          "id": "[CLS]",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "SpecialToken": {
+          "id": "[SEP]",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "B",
+          "type_id": 1
+        }
+      },
+      {
+        "SpecialToken": {
+          "id": "[SEP]",
+          "type_id": 1
+        }
+      }
+    ],
+    "special_tokens": {
+      "[CLS]": {
+        "id": "[CLS]",
+        "ids": [
+          2
+        ],
+        "tokens": [
+          "[CLS]"
+        ]
+      },
+      "[SEP]": {
+        "id": "[SEP]",
+        "ids": [
+          3
+        ],
+        "tokens": [
+          "[SEP]"
+        ]
+      }
+    }
+  },
+  "decoder": null,
+  "model": {
+    "type": "BPE",
+    "dropout": null,
+    "unk_token": "[UNK]",
+    "continuing_subword_prefix": null,
+    "end_of_word_suffix": null,
+    "fuse_unk": false,
+    "byte_fallback": false,
+    "vocab": {
+      "[PAD]": 0,
+      "[UNK]": 1,
+      "[CLS]": 2,
+      "[SEP]": 3,
+      "[MASK]": 4,
+      "A": 5,
+      "C": 6,
+      "G": 7,
+      "T": 8,
+      "TT": 9,
+      "AA": 10,
+      "TG": 11,
+      "AG": 12,
+      "CC": 13,
+      "TC": 14,
+      "AC": 15,
+      "GG": 16,
+      "ATT": 17,
+      "AT": 18,
+      "ATG": 19,
+      "GC": 20,
+      "TAA": 21,
+      "TCC": 22,
+      "ACC": 23,
+      "AAAA": 24,
+      "AGG": 25,
+      "ATC": 26,
+      "AGC": 27,
+      "TTC": 28,
+      "AAG": 29,
+      "TTTT": 30,
+      "TGC": 31,
+      "TGG": 32,
+      "AAC": 33,
+      "TTG": 34,
+      "TAG": 35,
+      "TAC": 36,
+      "CCC": 37,
+      "TATT": 38,
+      "TGGG": 39,
+      "TAT": 40,
+      "AGAA": 41,
+      "AGGG": 42,
+      "TTTC": 43,
+      "AGGC": 44,
+      "AGCC": 45,
+      "ATAA": 46,
+      "TGTG": 47,
+      "TTGG": 48,
+      "ATTC": 49,
+      "AAGG": 50,
+      "ACAC": 51,
+      "TCCC": 52,
+      "TCTC": 53,
+      "TATG": 54,
+      "TTTG": 55,
+      "TTCC": 56,
+      "AGTG": 57,
+      "ATGG": 58,
+      "AGAC": 59,
+      "AAAC": 60,
+      "ACCC": 61,
+      "TGCC": 62,
+      "ATTG": 63,
+      "ATCC": 64,
+      "AGAG": 65,
+      "ATGC": 66,
+      "ATAC": 67,
+      "TCTG": 68,
+      "TTAA": 69,
+      "TCAC": 70,
+      "TGAA": 71,
+      "TGGC": 72,
+      "TTGC": 73,
+      "TAAG": 74,
+      "TATC": 75,
+      "TAAC": 76,
+      "AAAG": 77,
+      "TTAC": 78,
+      "AAGC": 79,
+      "GGG": 80,
+      "TAGC": 81,
+      "GGC": 82,
+      "ATAT": 83,
+      "TACC": 84,
+      "AACC": 85,
+      "AATG": 86,
+      "TAGG": 87,
+      "GCC": 88,
+      "ATATT": 89,
+      "AGTC": 90,
+      "TTTTC": 91,
+      "AAAAC": 92,
+      "TGAC": 93,
+      "TTTAA": 94,
+      "AAAAG": 95,
+      "AATC": 96,
+      "TGTC": 97,
+      "TTATT": 98,
+      "ATAG": 99,
+      "TGAG": 100,
+      "TTTTG": 101,
+      "AAATT": 102,
+      "AATT": 103,
+      "AATAA": 104,
+      "TTTCC": 105,
+      "ACAG": 106,
+      "TCAG": 107,
+      "AAATG": 108,
+      "TGGGC": 109,
+      "ACTC": 110,
+      "AGGCC": 111,
+      "TTAG": 112,
+      "ACTG": 113,
+      "ACG": 114,
+      "ATATG": 115,
+      "TGGCC": 116,
+      "ATTTC": 117,
+      "ACAA": 118,
+      "ATCTC": 119,
+      "TATTC": 120,
+      "TGTAA": 121,
+      "ACTT": 122,
+      "ATGCC": 123,
+      "TAAAA": 124,
+      "AAAAAAAA": 125,
+      "ATTCC": 126,
+      "TTTAG": 127,
+      "TCCCC": 128,
+      "TTTGC": 129,
+      "TTCCC": 130,
+      "TGGGG": 131,
+      "TTCTC": 132,
+      "ATAAAA": 133,
+      "AGAAG": 134,
+      "TTTTTTTT": 135,
+      "ACCCC": 136,
+      "AGGGC": 137,
+      "ACCTC": 138,
+      "AGATG": 139,
+      "ATTAC": 140,
+      "AAGCC": 141,
+      "GGCC": 142,
+      "AGGAG": 143,
+      "TCAA": 144,
+      "ATTGC": 145,
+      "TATTG": 146,
+      "ATAAC": 147,
+      "ATATC": 148,
+      "TTTAC": 149,
+      "ATGGC": 150,
+      "AAGGC": 151,
+      "ACCAC": 152,
+      "GTG": 153,
+      "ATCCC": 154,
+      "AGAAC": 155,
+      "ATTTT": 156,
+      "TTGCC": 157,
+      "AAATC": 158,
+      "ATAAG": 159,
+      "TTGGC": 160,
+      "TGGAG": 161,
+      "ATGGG": 162,
+      "AAAGC": 163,
+      "AGGGG": 164,
+      "ATCAC": 165,
+      "ATTTG": 166,
+      "AATTC": 167,
+      "TGCAC": 168,
+      "TTTGG": 169,
+      "TCG": 170,
+      "AGAGC": 171,
+      "AAAGG": 172,
+      "GGGC": 173,
+      "TTGGG": 174,
+      "AGAAAA": 175,
+      "TATCC": 176,
+      "TCTCC": 177,
+      "ATAGC": 178,
+      "TGAGG": 179,
+      "TTTATT": 180,
+      "AGTAA": 181,
+      "AGAGG": 182,
+      "TCTTC": 183,
+      "ACATT": 184,
+      "TCCTG": 185,
+      "AGCCC": 186,
+      "TATGC": 187,
+      "TTAAAA": 188,
+      "AGATT": 189,
+      "TTAAC": 190,
+      "GGGG": 191,
+      "AAGAC": 192,
+      "TCATT": 193,
+      "TTCTG": 194,
+      "AGACC": 195,
+      "AAGGG": 196,
+      "ATACC": 197,
+      "TTTAT": 198,
+      "AAGTG": 199,
+      "TTATG": 200,
+      "AAGAA": 201,
+      "TAGCC": 202,
+      "TTCAC": 203,
+      "AGGTG": 204,
+      "TTGAA": 205,
+      "ATCTG": 206,
+      "AGCAC": 207,
+      "TGCTG": 208,
+      "AAACC": 209,
+      "ATGTG": 210,
+      "TTTTCC": 211,
+      "AGTTC": 212,
+      "TCCTC": 213,
+      "TATGG": 214,
+      "AATAC": 215,
+      "AGTGG": 216,
+      "TAGGC": 217,
+      "AGCTC": 218,
+      "ATAGG": 219,
+      "TTATC": 220,
+      "TTAAG": 221,
+      "TACCC": 222,
+      "TTTTTG": 223,
+      "AACAC": 224,
+      "TGCTC": 225,
+      "AGATC": 226,
+      "TCCCAGC": 227,
+      "AGCTG": 228,
+      "AATAG": 229,
+      "TCTTG": 230,
+      "AGTGGC": 231,
+      "ATTGG": 232,
+      "TACTC": 233,
+      "TAAAC": 234,
+      "AATGG": 235,
+      "AGGTC": 236,
+      "AGGAC": 237,
+      "TTGTG": 238,
+      "TATAC": 239,
+      "ATTTTC": 240,
+      "ATATAA": 241,
+      "AGGCTG": 242,
+      "ATTTAA": 243,
+      "AGTT": 244,
+      "AGTAG": 245,
+      "ATGAC": 246,
+      "AATGC": 247,
+      "TCCAC": 248,
+      "CCCC": 249,
+      "ATGTC": 250,
+      "AACTC": 251,
+      "TTTTTC": 252,
+      "TAAGC": 253,
+      "AAGTC": 254,
+      "TGGTG": 255,
+      "TATAA": 256,
+      "AGTGC": 257,
+      "TAAGG": 258,
+      "ACCTG": 259,
+      "TTAGC": 260,
+      "AAATAA": 261,
+      "TGCCTC": 262,
+      "AATCC": 263,
+      "TTGGCC": 264,
+      "TAGGG": 265,
+      "TGGAC": 266,
+      "TTGTC": 267,
+      "AACCC": 268,
+      "TTACC": 269,
+      "TAACC": 270,
+      "AATTTT": 271,
+      "AAAGAA": 272,
+      "ATTATT": 273,
+      "AGCG": 274,
+      "AAAAAC": 275,
+      "TAATG": 276,
+      "TTGAC": 277,
+      "AGTCC": 278,
+      "AACTG": 279,
+      "AGTTG": 280,
+      "AATTG": 281,
+      "TCTGC": 282,
+      "TTAGG": 283,
+      "TACAC": 284,
+      "AGAAGG": 285,
+      "ATATTC": 286,
+      "AAAACC": 287,
+      "AAAAGC": 288,
+      "TGCCC": 289,
+      "ACTGC": 290,
+      "AGAAGC": 291,
+      "TAATAA": 292,
+      "AATATT": 293,
+      "ACCATG": 294,
+      "TGGTC": 295,
+      "TTTTGC": 296,
+      "AACG": 297,
+      "TACTG": 298,
+      "ACACACAC": 299,
+      "ATTTTG": 300,
+      "TCCG": 301,
+      "TGCG": 302,
+      "AAAATG": 303,
+      "ACATG": 304,
+      "TCAGC": 305,
+      "ATCG": 306,
+      "AGTAC": 307,
+      "TTTTGG": 308,
+      "AATAT": 309,
+      "AGAGAA": 310,
+      "TTCG": 311,
+      "TCCAGCC": 312,
+      "ATATAC": 313,
+      "TCACC": 314,
+      "AAAAGG": 315,
+      "TGTGTGTG": 316,
+      "TCATC": 317,
+      "TGCTGGG": 318,
+      "TGAAG": 319,
+      "TGTAG": 320,
+      "TGTGG": 321,
+      "AAAAATT": 322,
+      "ACTTC": 323,
+      "TTCCCC": 324,
+      "ATAGAA": 325,
+      "TTGCCC": 326,
+      "AGGAGG": 327,
+      "TTTCCC": 328,
+      "TATATT": 329,
+      "ACCG": 330,
+      "ACTAC": 331,
+      "TCACTGC": 332,
+      "GCG": 333,
+      "TTTGTG": 334,
+      "ACAGC": 335,
+      "TCATG": 336,
+      "AGTTTT": 337,
+      "AGGAA": 338,
+      "TTTATG": 339,
+      "ATATTG": 340,
+      "TGATG": 341,
+      "TCTAA": 342,
+      "TGTGC": 343,
+      "AGGAAG": 344,
+      "TTTGGG": 345,
+      "TGTTC": 346,
+      "AGCCCC": 347,
+      "AGTTTC": 348,
+      "AGGCTGG": 349,
+      "TTTGCC": 350,
+      "ATTTCC": 351,
+      "ATACAC": 352,
+      "AAAATAA": 353,
+      "TAGAC": 354,
+      "AGGAGAA": 355,
+      "TGAGC": 356,
+      "TGGAA": 357,
+      "TTTTTAA": 358,
+      "AGCCTCCC": 359,
+      "ATGAA": 360,
+      "TTTAAG": 361,
+      "TCTGG": 362,
+      "TTTATC": 363,
+      "TTATAA": 364,
+      "TGATT": 365,
+      "AACAA": 366,
+      "TAGCTGGG": 367,
+      "TCAAG": 368,
+      "AAAAAA": 369,
+      "ACTTTGGG": 370,
+      "TATTCC": 371,
+      "TCAGG": 372,
+      "AACAG": 373,
+      "TTCTTC": 374,
+      "TGTGGC": 375,
+      "ATATGC": 376,
+      "ATTACAGGC": 377,
+      "AGGGGC": 378,
+      "AGGGCC": 379,
+      "TTATTC": 380,
+      "ATATCC": 381,
+      "TGTAATCCCAGC": 382,
+      "TACG": 383,
+      "AGAAAC": 384,
+      "TGTCC": 385,
+      "AGATGG": 386,
+      "TGTGCC": 387,
+      "TTTCTC": 388,
+      "TGAAC": 389,
+      "AGTCTC": 390,
+      "TGTTG": 391,
+      "ATTTTTT": 392,
+      "AAGAAG": 393,
+      "TGGGGC": 394,
+      "AGCAGC": 395,
+      "GCCC": 396,
+      "TTTGGC": 397,
+      "AGGCTGAGGC": 398,
+      "TGGGCC": 399,
+      "TTCTCC": 400,
+      "TAGAA": 401,
+      "TGGAGTGC": 402,
+      "ATTAA": 403,
+      "AGTGCC": 404,
+      "TGTCTC": 405,
+      "ATATGG": 406,
+      "ACATC": 407,
+      "TGGGGG": 408,
+      "TGACC": 409,
+      "ACTCC": 410,
+      "TAAAAC": 411,
+      "AGATAA": 412,
+      "TAATTTT": 413,
+      "TCAAC": 414,
+      "TCTAC": 415,
+      "TCTAG": 416,
+      "GAG": 417,
+      "TAAATG": 418,
+      "AGCAA": 419,
+      "TATATG": 420,
+      "ATATATAT": 421,
+      "ATTTGC": 422,
+      "TCCTCC": 423,
+      "CCCAC": 424,
+      "ATTTATT": 425,
+      "TCTGCC": 426,
+      "ATGGCC": 427,
+      "TCGC": 428,
+      "AGTATT": 429,
+      "AGAACC": 430,
+      "TTAAAC": 431,
+      "AAATTC": 432,
+      "AGAGAC": 433,
+      "ATTTAC": 434,
+      "ATTGCC": 435,
+      "AACAAC": 436,
+      "TTTAAC": 437,
+      "ACGG": 438,
+      "AAGAAAA": 439,
+      "TCTGGC": 440,
+      "ATTCTCC": 441,
+      "AGGTGG": 442,
+      "TGCTGC": 443,
+      "TTCAAG": 444,
+      "AGAGGG": 445,
+      "ACACC": 446,
+      "TCTTTT": 447,
+      "AGAGGC": 448,
+      "ATCACC": 449,
+      "TAAATT": 450,
+      "AAGGCC": 451,
+      "TTGCAGTG": 452,
+      "TGTAC": 453,
+      "AATTTC": 454,
+      "ATCCCC": 455,
+      "ACAAG": 456,
+      "ACAGG": 457,
+      "ACAAC": 458,
+      "TGCCCC": 459,
+      "AGATTC": 460,
+      "TTAGAA": 461,
+      "TTGGGG": 462,
+      "AGACAC": 463,
+      "TGGAAG": 464,
+      "ACCTCC": 465,
+      "ATGGGG": 466,
+      "AGCCTCC": 467,
+      "TTATTG": 468,
+      "TAAAAG": 469,
+      "ATCTTC": 470,
+      "ATCTCC": 471,
+      "TGAAGC": 472,
+      "TAATC": 473,
+      "AAATGC": 474,
+      "TTGTTG": 475,
+      "ATTCCC": 476,
+      "TACTAAAA": 477,
+      "ATAGTG": 478,
+      "AAATAC": 479,
+      "TTGGGC": 480,
+      "TAGAGAC": 481,
+      "TGTTTT": 482,
+      "TTCTGC": 483,
+      "TGGCCC": 484,
+      "TCTGTC": 485,
+      "AGCTCC": 486,
+      "AACTCC": 487,
+      "TTAGCC": 488,
+      "AAAGTGCTGGG": 489,
+      "ATAGAC": 490,
+      "TATTTTTAG": 491,
+      "ACTTG": 492,
+      "ACCACC": 493,
+      "AAACAC": 494,
+      "GTGG": 495,
+      "ATTTAG": 496,
+      "AGGAGC": 497,
+      "AGGCTGGAGTGC": 498,
+      "ATACCC": 499,
+      "ATGTAA": 500,
+      "ACGC": 501,
+      "AGTAT": 502,
+      "TTTACC": 503,
+      "ACTAA": 504,
+      "AGGCCC": 505,
+      "AAGGGG": 506,
+      "TCTCG": 507,
+      "ATGAAG": 508,
+      "AAAGAC": 509,
+      "TGAAAA": 510,
+      "AAGGGC": 511,
+      "ATAGGC": 512,
+      "AGAGTG": 513,
+      "AGCTGC": 514,
+      "ATGTTC": 515,
+      "TATTTC": 516,
+      "TGATC": 517,
+      "AGTTTG": 518,
+      "AGCTAA": 519,
+      "AGAGCC": 520,
+      "TGCTTC": 521,
+      "ATCATC": 522,
+      "AACATGG": 523,
+      "AGCTTC": 524,
+      "AAGAAC": 525,
+      "TTTTTTG": 526,
+      "AGGGGG": 527,
+      "ATAAGC": 528,
+      "TAAGCC": 529,
+      "ACTGG": 530,
+      "ACAAAA": 531,
+      "ATCATT": 532,
+      "TCTTTC": 533,
+      "ATGATG": 534,
+      "TGCAA": 535,
+      "AGGTTC": 536,
+      "AACATT": 537,
+      "ATGGGC": 538,
+      "ATAGAG": 539,
+      "AAATGG": 540,
+      "AGTTCC": 541,
+      "TTTAGC": 542,
+      "AACTTC": 543,
+      "AGCAAG": 544,
+      "ATAAAAC": 545,
+      "AAAATC": 546,
+      "AGCCAC": 547,
+      "AGGAAC": 548,
+      "TTAACC": 549,
+      "TATTTATT": 550,
+      "TTTCTG": 551,
+      "ATAAGG": 552,
+      "AGCCACC": 553,
+      "AGATGC": 554,
+      "TTAAGC": 555,
+      "TTGTAA": 556,
+      "AGTGTG": 557,
+      "AACCCC": 558,
+      "TTCATT": 559,
+      "ATCATG": 560,
+      "AATGAA": 561,
+      "AGGTGC": 562,
+      "AAAAAAAAAAAAAAAA": 563,
+      "AGGATG": 564,
+      "AGCCG": 565,
+      "TGGTGG": 566,
+      "AGTGGG": 567,
+      "TGCACTCCAGCC": 568,
+      "TATTGC": 569,
+      "TAGTC": 570,
+      "CCCG": 571,
+      "AAGTAA": 572,
+      "TAGTG": 573,
+      "TTTTTTTTTTTTTTTT": 574,
+      "AGCATT": 575,
+      "ATCTGC": 576,
+      "TCTCAC": 577,
+      "AAATTG": 578,
+      "TTTAGG": 579,
+      "AGACCC": 580,
+      "GGGCC": 581,
+      "TCCTTC": 582,
+      "ATAGGG": 583,
+      "AATATG": 584,
+      "TTATAC": 585,
+      "TAGAAG": 586,
+      "AAAGTG": 587,
+      "AAATCC": 588,
+      "TTCCTC": 589,
+      "TTTCAC": 590,
+      "AGTATG": 591,
+      "TACTAAAAATAC": 592,
+      "ATGTGC": 593,
+      "AGGAGGC": 594,
+      "TATATC": 595,
+      "TTCTAA": 596,
+      "TGAGGC": 597,
+      "ACACAC": 598,
+      "TCCCCC": 599,
+      "AACATC": 600,
+      "AAGCG": 601,
+      "AATGGC": 602,
+      "ACCCCC": 603,
+      "AGATAC": 604,
+      "ATAAAAG": 605,
+      "ATGATT": 606,
+      "TGGAGG": 607,
+      "AGTTAA": 608,
+      "": 609
+    },
+    "merges": [
+      "A ",
+      "C ",
+      "G ",
+      "T ",
+      "T T",
+      "A A",
+      "T G",
+      "A G",
+      "C C",
+      "T C",
+      "A C",
+      "G G",
+      "A TT",
+      "A T",
+      "A TG",
+      "G C",
+      "T AA",
+      "T CC",
+      "A CC",
+      "AA AA",
+      "AG G",
+      "A TC",
+      "AG C",
+      "TT C",
+      "AA G",
+      "TT TT",
+      "TG C",
+      "TG G",
+      "AA C",
+      "TT G",
+      "T AG",
+      "T AC",
+      "CC C",
+      "T ATT",
+      "TG GG",
+      "T AT",
+      "AG AA",
+      "AG GG",
+      "TT TC",
+      "AG GC",
+      "AG CC",
+      "AT AA",
+      "TG TG",
+      "TT GG",
+      "ATT C",
+      "AA GG",
+      "AC AC",
+      "TCC C",
+      "TC TC",
+      "T ATG",
+      "TT TG",
+      "TT CC",
+      "AG TG",
+      "ATG G",
+      "AG AC",
+      "AA AC",
+      "ACC C",
+      "TG CC",
+      "ATT G",
+      "AT CC",
+      "AG AG",
+      "ATG C",
+      "AT AC",
+      "TC TG",
+      "TT AA",
+      "TC AC",
+      "TG AA",
+      "TG GC",
+      "TT GC",
+      "TAA G",
+      "T ATC",
+      "TAA C",
+      "AA AG",
+      "TT AC",
+      "AA GC",
+      "GG G",
+      "T AGC",
+      "GG C",
+      "AT AT",
+      "T ACC",
+      "AA CC",
+      "AA TG",
+      "T AGG",
+      "G CC",
+      "AT ATT",
+      "AG TC",
+      "TT TTC",
+      "AAAA C",
+      "TG AC",
+      "TT TAA",
+      "AAAA G",
+      "AA TC",
+      "TG TC",
+      "TT ATT",
+      "AT AG",
+      "TG AG",
+      "TTTT G",
+      "AA ATT",
+      "AA TT",
+      "AA TAA",
+      "TT TCC",
+      "AC AG",
+      "TC AG",
+      "AA ATG",
+      "TGGG C",
+      "AC TC",
+      "AGG CC",
+      "TT AG",
+      "AC TG",
+      "AC G",
+      "AT ATG",
+      "TGG CC",
+      "ATT TC",
+      "AC AA",
+      "ATC TC",
+      "TATT C",
+      "TG TAA",
+      "AC TT",
+      "ATG CC",
+      "TAA AA",
+      "AAAA AAAA",
+      "ATT CC",
+      "TT TAG",
+      "TCC CC",
+      "TT TGC",
+      "TT CCC",
+      "TGGG G",
+      "TTC TC",
+      "AT AAAA",
+      "AG AAG",
+      "TTTT TTTT",
+      "ACC CC",
+      "AGGG C",
+      "ACC TC",
+      "AG ATG",
+      "ATT AC",
+      "AAG CC",
+      "GG CC",
+      "AGG AG",
+      "TC AA",
+      "ATT GC",
+      "TATT G",
+      "AT AAC",
+      "AT ATC",
+      "TT TAC",
+      "ATG GC",
+      "AAGG C",
+      "ACC AC",
+      "G TG",
+      "AT CCC",
+      "AG AAC",
+      "ATT TT",
+      "TTG CC",
+      "AA ATC",
+      "AT AAG",
+      "TTGG C",
+      "TGG AG",
+      "ATG GG",
+      "AA AGC",
+      "AGGG G",
+      "ATC AC",
+      "ATT TG",
+      "AA TTC",
+      "TGC AC",
+      "TT TGG",
+      "TC G",
+      "AG AGC",
+      "AA AGG",
+      "GG GC",
+      "TTGG G",
+      "AG AAAA",
+      "TAT CC",
+      "TC TCC",
+      "AT AGC",
+      "TG AGG",
+      "TT TATT",
+      "AG TAA",
+      "AG AGG",
+      "TC TTC",
+      "AC ATT",
+      "TCC TG",
+      "AG CCC",
+      "TATG C",
+      "TT AAAA",
+      "AG ATT",
+      "TT AAC",
+      "GG GG",
+      "AAG AC",
+      "TC ATT",
+      "TTC TG",
+      "AG ACC",
+      "AAGG G",
+      "AT ACC",
+      "TT TAT",
+      "AAG TG",
+      "TT ATG",
+      "AAG AA",
+      "TAG CC",
+      "TTC AC",
+      "AGG TG",
+      "TTG AA",
+      "ATC TG",
+      "AGC AC",
+      "TGC TG",
+      "AA ACC",
+      "ATG TG",
+      "TTTT CC",
+      "AG TTC",
+      "TCC TC",
+      "TATG G",
+      "AA TAC",
+      "AG TGG",
+      "TAG GC",
+      "AGC TC",
+      "AT AGG",
+      "TT ATC",
+      "TT AAG",
+      "T ACCC",
+      "TTTT TG",
+      "AAC AC",
+      "TGC TC",
+      "AG ATC",
+      "TCCC AGC",
+      "AGC TG",
+      "AA TAG",
+      "TC TTG",
+      "AGTG GC",
+      "ATT GG",
+      "TAC TC",
+      "TAA AC",
+      "AA TGG",
+      "AGG TC",
+      "AGG AC",
+      "TTG TG",
+      "TAT AC",
+      "ATT TTC",
+      "AT ATAA",
+      "AGGC TG",
+      "ATT TAA",
+      "AG TT",
+      "AG TAG",
+      "ATG AC",
+      "AA TGC",
+      "TCC AC",
+      "CC CC",
+      "ATG TC",
+      "AAC TC",
+      "TTTT TC",
+      "TAA GC",
+      "AAG TC",
+      "TGG TG",
+      "TAT AA",
+      "AG TGC",
+      "TAA GG",
+      "ACC TG",
+      "TT AGC",
+      "AA ATAA",
+      "TGCC TC",
+      "AA TCC",
+      "TTGG CC",
+      "TAG GG",
+      "TGG AC",
+      "TTG TC",
+      "AA CCC",
+      "TT ACC",
+      "TAA CC",
+      "AA TTTT",
+      "AA AGAA",
+      "ATT ATT",
+      "AGC G",
+      "AAAA AC",
+      "TAA TG",
+      "TTG AC",
+      "AG TCC",
+      "AAC TG",
+      "AG TTG",
+      "AA TTG",
+      "TC TGC",
+      "TT AGG",
+      "TAC AC",
+      "AGAA GG",
+      "AT ATTC",
+      "AAAA CC",
+      "AAAA GC",
+      "TG CCC",
+      "AC TGC",
+      "AGAA GC",
+      "TAA TAA",
+      "AA TATT",
+      "ACC ATG",
+      "TGG TC",
+      "TTTT GC",
+      "AAC G",
+      "TAC TG",
+      "ACAC ACAC",
+      "ATT TTG",
+      "TCC G",
+      "TGC G",
+      "AAAA TG",
+      "AC ATG",
+      "TC AGC",
+      "ATC G",
+      "AG TAC",
+      "TTTT GG",
+      "AA TAT",
+      "AG AGAA",
+      "TTC G",
+      "TCC AGCC",
+      "AT ATAC",
+      "TC ACC",
+      "AAAA GG",
+      "TGTG TGTG",
+      "TC ATC",
+      "TGC TGGG",
+      "TG AAG",
+      "TG TAG",
+      "TG TGG",
+      "AAAA ATT",
+      "AC TTC",
+      "TTCC CC",
+      "AT AGAA",
+      "TTG CCC",
+      "AGG AGG",
+      "TT TCCC",
+      "TAT ATT",
+      "ACC G",
+      "AC TAC",
+      "TCAC TGC",
+      "GC G",
+      "TT TGTG",
+      "AC AGC",
+      "TC ATG",
+      "AG TTTT",
+      "AGG AA",
+      "TT TATG",
+      "AT ATTG",
+      "TG ATG",
+      "TC TAA",
+      "TG TGC",
+      "AGG AAG",
+      "TT TGGG",
+      "TG TTC",
+      "AGCC CC",
+      "AG TTTC",
+      "AGGC TGG",
+      "TTTG CC",
+      "ATT TCC",
+      "AT ACAC",
+      "AAAA TAA",
+      "TAG AC",
+      "AGG AGAA",
+      "TG AGC",
+      "TGG AA",
+      "TTTT TAA",
+      "AGCC TCCC",
+      "ATG AA",
+      "TT TAAG",
+      "TC TGG",
+      "TT TATC",
+      "TT ATAA",
+      "TG ATT",
+      "AAC AA",
+      "TAGC TGGG",
+      "TC AAG",
+      "AAAA AA",
+      "ACTT TGGG",
+      "TATT CC",
+      "TC AGG",
+      "AAC AG",
+      "TTC TTC",
+      "TGTG GC",
+      "AT ATGC",
+      "ATTAC AGGC",
+      "AGGG GC",
+      "AGGG CC",
+      "TT ATTC",
+      "AT ATCC",
+      "TGTAA TCCCAGC",
+      "TAC G",
+      "AGAA AC",
+      "TG TCC",
+      "AG ATGG",
+      "TGTG CC",
+      "TTTC TC",
+      "TG AAC",
+      "AG TCTC",
+      "TG TTG",
+      "ATT TTTT",
+      "AAG AAG",
+      "TGGG GC",
+      "AGC AGC",
+      "G CCC",
+      "TTTG GC",
+      "AGGCTG AGGC",
+      "TGGG CC",
+      "TTC TCC",
+      "TAG AA",
+      "TGGAG TGC",
+      "ATT AA",
+      "AGTG CC",
+      "TG TCTC",
+      "AT ATGG",
+      "AC ATC",
+      "TGGG GG",
+      "TG ACC",
+      "AC TCC",
+      "TAA AAC",
+      "AG ATAA",
+      "TAA TTTT",
+      "TC AAC",
+      "TC TAC",
+      "TC TAG",
+      "G AG",
+      "TAA ATG",
+      "AGC AA",
+      "TAT ATG",
+      "ATAT ATAT",
+      "ATT TGC",
+      "TCC TCC",
+      "CCC AC",
+      "ATT TATT",
+      "TC TGCC",
+      "ATGG CC",
+      "TC GC",
+      "AG TATT",
+      "AGAA CC",
+      "TT AAAC",
+      "AA ATTC",
+      "AG AGAC",
+      "ATT TAC",
+      "ATTG CC",
+      "AAC AAC",
+      "TT TAAC",
+      "AC GG",
+      "AAG AAAA",
+      "TCTG GC",
+      "ATTC TCC",
+      "AGG TGG",
+      "TGC TGC",
+      "TTC AAG",
+      "AG AGGG",
+      "AC ACC",
+      "TC TTTT",
+      "AG AGGC",
+      "ATC ACC",
+      "TAA ATT",
+      "AAGG CC",
+      "TTGC AGTG",
+      "TG TAC",
+      "AA TTTC",
+      "ATCC CC",
+      "AC AAG",
+      "AC AGG",
+      "AC AAC",
+      "TGCC CC",
+      "AG ATTC",
+      "TT AGAA",
+      "TTGG GG",
+      "AG ACAC",
+      "TGG AAG",
+      "ACC TCC",
+      "ATG GGG",
+      "AGCC TCC",
+      "TT ATTG",
+      "TAA AAG",
+      "ATC TTC",
+      "ATC TCC",
+      "TGAA GC",
+      "TAA TC",
+      "AA ATGC",
+      "TTG TTG",
+      "ATT CCC",
+      "TAC TAAAA",
+      "AT AGTG",
+      "AA ATAC",
+      "TTGG GC",
+      "TAG AGAC",
+      "TG TTTT",
+      "TTC TGC",
+      "TGG CCC",
+      "TCTG TC",
+      "AGC TCC",
+      "AAC TCC",
+      "TT AGCC",
+      "AAAG TGCTGGG",
+      "AT AGAC",
+      "TATT TTTAG",
+      "AC TTG",
+      "ACC ACC",
+      "AA ACAC",
+      "G TGG",
+      "ATT TAG",
+      "AGG AGC",
+      "AGGC TGGAGTGC",
+      "AT ACCC",
+      "ATG TAA",
+      "AC GC",
+      "AG TAT",
+      "TT TACC",
+      "AC TAA",
+      "AGG CCC",
+      "AAGG GG",
+      "TCTC G",
+      "ATG AAG",
+      "AA AGAC",
+      "TG AAAA",
+      "AAGG GC",
+      "AT AGGC",
+      "AG AGTG",
+      "AGC TGC",
+      "ATG TTC",
+      "TATT TC",
+      "TG ATC",
+      "AG TTTG",
+      "AGC TAA",
+      "AG AGCC",
+      "TGC TTC",
+      "ATC ATC",
+      "AAC ATGG",
+      "AGC TTC",
+      "AAG AAC",
+      "TTTT TTG",
+      "AGGG GG",
+      "ATAA GC",
+      "TAAG CC",
+      "AC TGG",
+      "AC AAAA",
+      "ATC ATT",
+      "TC TTTC",
+      "ATG ATG",
+      "TGC AA",
+      "AGG TTC",
+      "AAC ATT",
+      "ATG GGC",
+      "AT AGAG",
+      "AA ATGG",
+      "AG TTCC",
+      "TT TAGC",
+      "AAC TTC",
+      "AGC AAG",
+      "AT AAAAC",
+      "AAAA TC",
+      "AGCC AC",
+      "AGG AAC",
+      "TTAA CC",
+      "TATT TATT",
+      "TTTC TG",
+      "ATAA GG",
+      "AGCC ACC",
+      "AG ATGC",
+      "TTAA GC",
+      "TTG TAA",
+      "AG TGTG",
+      "AACC CC",
+      "TTC ATT",
+      "ATC ATG",
+      "AA TGAA",
+      "AGG TGC",
+      "AAAAAAAA AAAAAAAA",
+      "AGG ATG",
+      "AGCC G",
+      "TGG TGG",
+      "AG TGGG",
+      "TGCAC TCCAGCC",
+      "TATT GC",
+      "TAG TC",
+      "CCC G",
+      "AAG TAA",
+      "TAG TG",
+      "TTTTTTTT TTTTTTTT",
+      "AGC ATT",
+      "ATC TGC",
+      "TCTC AC",
+      "AA ATTG",
+      "TT TAGG",
+      "AG ACCC",
+      "GGG CC",
+      "TCC TTC",
+      "AT AGGG",
+      "AA TATG",
+      "TT ATAC",
+      "TAG AAG",
+      "AA AGTG",
+      "AA ATCC",
+      "TTCC TC",
+      "TTTC AC",
+      "AG TATG",
+      "TACTAAAA ATAC",
+      "ATG TGC",
+      "AGG AGGC",
+      "TAT ATC",
+      "TTC TAA",
+      "TG AGGC",
+      "ACAC AC",
+      "TCC CCC",
+      "AAC ATC",
+      "AAGC G",
+      "AA TGGC",
+      "ACC CCC",
+      "AG ATAC",
+      "AT AAAAG",
+      "ATG ATT",
+      "TGG AGG",
+      "AG TTAA"
+    ]
+  }
+}