fix: consolidate DeepCatch fixes — paper, docker, syntax, citation, gitignore, CI

.github/workflows/compile-paper.yml

@@ -4,7 +4,8 @@ on:
   push:
     paths:
       - 'paper/deepcatch_final.tex'
-      - 'paper/deepcatch_references.bib'
+      - 'paper/supplementary.tex'
+      - 'paper/references_final.bib'
       - 'paper/figures/**'
   workflow_dispatch:
 

.gitignore

@@ -27,7 +27,9 @@ build/
 *.ipynb
 
 # ── Machine Learning ────────────────────────────────────────────────────────
-*.pt
+# Only ignore generated/trained checkpoints, not bundled ones
+model/*.pt
+!model/motif_model_checkpoint.pt
 *.pth
 *.h5
 *.ckpt

CITATION.cff

@@ -3,14 +3,14 @@ message: "If you use DeepCatch in your research, please cite it as follows."
 title: "DeepCatch: Performance-Weighted Multi-Modal Fusion for Ultra-Early Cancer Detection from cfDNA"
 authors:
   - given-names: "Royce"
-    family-names: "[Last Name]"
-    affiliation: "DeepCatch Consortium"
-    orcid: ""
+    family-names: "Lam"
+    affiliation: "DeepCatch Project"
+    orcid: "https://orcid.org/0009-0009-1904-6999"
   - name: "DeepCatch Contributors"
 year: 2026
 doi: "to be assigned (preprint)"
-url: "https://github.com/deepcatch/deepcatch"
-repository-code: "https://github.com/deepcatch/deepcatch"
+url: "https://github.com/rollroyces/deepcatch"
+repository-code: "https://github.com/rollroyces/deepcatch"
 license: MIT
 version: "1.0.0-preprint"
 date-released: 2026-04-28

Dockerfile

@@ -53,18 +53,16 @@ RUN pip install --no-cache-dir \
     pyyaml==6.0.1
 
 # ── Copy source code ─────────────────────────────────────────────────────
-COPY validation_framework.py /app/
 COPY validation/ /app/validation/
-COPY reproducibility/ /app/reproducibility/
 COPY run_bioinfo_validation.py /app/
+COPY run_full_validation.py /app/
 
 # ── Create output directory ──────────────────────────────────────────────
 RUN mkdir -p /app/results
 
 # ── Verify imports ───────────────────────────────────────────────────────
-RUN python -c "import validation_framework; print('validation_framework OK')" && \
-    python -c "from validation import *; print('validation modules OK')"
+RUN python -c "from validation import *; print('validation modules OK')"
 
 # ── Entry point ──────────────────────────────────────────────────────────
-# Default: run ALL validation modules
-CMD ["python", "run_bioinfo_validation.py"]
+# Default: run full validation suite
+CMD ["python", "run_full_validation.py"]

RUN_ALL.sh

@@ -134,7 +134,7 @@ echo ""
 # Step 1: Install dependencies (if needed)
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[1/6] Installing dependencies...${NC}"
+echo -e "${BLUE}${BOLD}[1/5] Installing dependencies...${NC}"
 
 if [ -f "requirements.txt" ]; then
     $PYTHON -m pip install -r requirements.txt --quiet 2>/dev/null || {
@@ -150,7 +150,7 @@ echo ""
 # Step 2: Full cross-validation
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[2/6] Running full cross-validation...${NC}"
+echo -e "${BLUE}${BOLD}[2/5] Running full cross-validation...${NC}"
 echo ""
 
 $PYTHON run_full_validation.py \
@@ -170,65 +170,65 @@ echo -e "${GREEN}[OK] Full validation complete${NC}"
 echo ""
 
 # ------------------------------------------------------------------
-# Step 3: TCGA validation
+# Step 3: Real data validation (TCGA) — optional
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[3/6] Running TCGA real data validation...${NC}"
+echo -e "${BLUE}${BOLD}[3/5] Running real data validation...${NC}"
 echo ""
 
-$PYTHON run_tcga_validation.py \
-    $QUICK_FLAG \
-    --output results/ \
-    --seeds 5
-
-TCGA_EXIT=$?
-if [ $TCGA_EXIT -ne 0 ]; then
-    echo -e "${YELLOW}[WARN] TCGA validation exited with code $TCGA_EXIT (non-fatal)${NC}"
+TCGA_SCRIPT="$(dirname $0)/real_tcga_validation.py"
+if [ -f "$TCGA_SCRIPT" ]; then
+    $PYTHON real_tcga_validation.py \
+        $QUICK_FLAG \
+        --output results/ \
+        --seeds 5 || echo -e "${YELLOW}[WARN] TCGA validation non-fatal error${NC}"
+else
+    echo -e "${YELLOW}[WARN] real_tcga_validation.py not found — skipping${NC}"
 fi
 
 echo ""
-echo -e "${GREEN}[OK] TCGA validation complete${NC}"
+echo -e "${GREEN}[OK] TCGA validation step complete${NC}"
 echo ""
 
 # ------------------------------------------------------------------
-# Step 4: Benchmark comparison
+# Step 4: Jiang 4-mer analysis — optional
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[4/6] Generating benchmark comparison...${NC}"
+echo -e "${BLUE}${BOLD}[4/5] Running Jiang 4-mer analysis...${NC}"
 echo ""
 
-$PYTHON run_benchmark_comparison.py \
-    --results results/final_cross_validated_results.json \
-    --tcga results/tcga_validation_results.json \
-    --output results/
-
-BENCH_EXIT=$?
-if [ $BENCH_EXIT -ne 0 ]; then
-    echo -e "${YELLOW}[WARN] Benchmark comparison exited with code $BENCH_EXIT (non-fatal)${NC}"
+JIANG_SCRIPT="$(dirname $0)/run_jiang_analysis.py"
+if [ -f "$JIANG_SCRIPT" ]; then
+    $PYTHON run_jiang_analysis.py \
+        $QUICK_FLAG \
+        --output results/ || echo -e "${YELLOW}[WARN] Jiang analysis non-fatal error${NC}"
+else
+    echo -e "${YELLOW}[WARN] run_jiang_analysis.py not found — skipping${NC}"
 fi
 
 echo ""
-echo -e "${GREEN}[OK] Benchmark comparison complete${NC}"
+echo -e "${GREEN}[OK] Jiang analysis step complete${NC}"
 echo ""
 
 # ------------------------------------------------------------------
-# Step 5: Verify fixes
+# Step 5: Supplementary checks
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[5/6] Running fix verification...${NC}"
+echo -e "${BLUE}${BOLD}[5/5] Running supplementary checks...${NC}"
 echo ""
 
-if [ -f "validation_framework_tests.py" ]; then
-    $PYTHON validation_framework_tests.py 2>&1 | head -30 || true
-    echo ""
-fi
-
-if [ -f "verify_fixes.py" ]; then
-    $PYTHON verify_fixes.py 2>&1 || {
-        echo -e "${YELLOW}[WARN] verify_fixes.py issue (non-fatal)${NC}"
+# Verify Python syntax across all .py files
+BROKEN=0
+for f in $(find . -name '*.py' -not -path './.git/*' | sort); do
+    $PYTHON -c "import ast; ast.parse(open('$f').read())" 2>/dev/null || {
+        echo -e "  ${RED}⚠ Syntax error in $f${NC}"
+        BROKEN=$((BROKEN+1))
     }
+done
+if [ $BROKEN -eq 0 ]; then
+    echo -e "  ${GREEN}✓ All Python files pass syntax check${NC}"
 else
-    echo -e "${YELLOW}[WARN] verify_fixes.py not found — skipping${NC}"
+    echo -e "  ${YELLOW}⚠ $BROKEN file(s) have syntax errors${NC}"
 fi
 
 echo ""
@@ -237,7 +237,7 @@ echo ""
 # Step 6: Summary
 # ------------------------------------------------------------------
 
-echo -e "${BLUE}${BOLD}[6/6] Generating summary...${NC}"
+echo -e "${BLUE}${BOLD}[Summary]${NC}"
 echo ""
 
 echo -e "${BLUE}${BOLD}========================================${NC}"

paper/README.md

@@ -6,35 +6,37 @@ This directory contains the LaTeX source for the DeepCatch manuscript.
 
 | File | Description |
 |------|-------------|
-| `main.tex` | Primary manuscript (Nature Medicine / Lancet Oncology / Cancer Discovery target) |
-| `references.bib` | Bibliography in BibTeX format |
+| `deepcatch_final.tex` | Primary manuscript (Bioinformatics target) |
+| `references_final.bib` | Bibliography in BibTeX format |
 | `supplementary.tex` | Supplementary materials (extended methods, additional figures) |
-| `abstract.txt` | Plain-text abstract for submission systems |
-| `cover_letter.txt` | Template cover letter for journal submission |
-| `figures/` | Figure files (PNG, PDF, or TikZ sources) |
+| `figures/` | Figure files (PNG, PDF, or TikZ sources)
 
 ## How to Compile
 
 ### Option 1: pdflatex + bibtex (traditional)
 
 ```bash
 cd paper/
-pdflatex main.tex
-bibtex main
-pdflatex main.tex
-pdflatex main.tex
+pdflatex deepcatch_final.tex
+bibtex deepcatch_final
+pdflatex deepcatch_final.tex
+pdflatex deepcatch_final.tex
 ```
 
 ### Option 2: latexmk (recommended)
 
 ```bash
 cd paper/
-latexmk -pdf main.tex
+latexmk -pdf deepcatch_final.tex
 ```
 
 ### Option 3: Overleaf
 
-Upload the entire `paper/` directory to Overleaf. Set the main document to `main.tex`.
+Upload the entire `paper/` directory to Overleaf. Set the main document to `deepcatch_final.tex`.
+
+### Option 4: GitHub CI
+
+Push changes to `paper/deepcatch_final.tex` or `paper/supplementary.tex` — the CI workflow will automatically compile and upload the PDF as an artifact.
 
 ## Dependencies