#!/usr/bin/env python3


import pandas as pd
import numpy as np

# Create a sample count matrix (genes x samples)
np.random.seed(42)
genes = ["BRCA1", "TP53", "EGFR", "MYC", "KRAS", "BRAF", "PIK3CA", "PTEN",
         "RB1", "APC", "VHL", "WT1", "NF1", "RET", "ALK"]

# Control samples (3 replicates)
control = np.random.poisson(lam=100, size=(len(genes), 3))
# Treatment samples -- some genes upregulated, some down
treatment = np.random.poisson(lam=100, size=(len(genes), 3))
# Simulate differential expression
treatment[0] *= 3   # BRCA1 upregulated
treatment[1] *= 4   # TP53 upregulated
treatment[4] //= 3  # KRAS downregulated
treatment[7] //= 2  # PTEN downregulated

df = pd.DataFrame(
    np.hstack([control, treatment]),
    index=genes,
    columns=["ctrl_1", "ctrl_2", "ctrl_3", "treat_1", "treat_2", "treat_3"]
)

print("Gene expression count matrix:")
print(df)
print(f"\nShape: {df.shape[0]} genes x {df.shape[1]} samples")

# Calculate means
df["ctrl_mean"] = df[["ctrl_1", "ctrl_2", "ctrl_3"]].mean(axis=1)
df["treat_mean"] = df[["treat_1", "treat_2", "treat_3"]].mean(axis=1)
df["fold_change"] = df["treat_mean"] / df["ctrl_mean"]
df["log2_fc"] = np.log2(df["fold_change"].replace(0, np.nan))

print("\nDifferential expression summary:")
print("-" * 60)
summary = df[["ctrl_mean", "treat_mean", "fold_change", "log2_fc"]].round(2)
print(summary.sort_values("log2_fc", ascending=False))

# Classify
print("\nUpregulated (log2FC > 1):")
up = summary[summary["log2_fc"] > 1]
for gene in up.index:
    print(f"  {gene}: log2FC = {up.loc[gene, 'log2_fc']}")

print("\nDownregulated (log2FC < -1):")
down = summary[summary["log2_fc"] < -1]
for gene in down.index:
    print(f"  {gene}: log2FC = {down.loc[gene, 'log2_fc']}")

#!/usr/bin/env python3


import os
import pandas as pd
import numpy as np

try:
    from openai import OpenAI
    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", ""), base_url=os.environ.get("OPENAI_BASE_URL", "https://openrouter.ai/api/v1"))
    AI_AVAILABLE = True
except ImportError:
    AI_AVAILABLE = False
    print("Note: Install openai package for AI features\n")

def ask_ai(prompt: str) -> str:
    if not AI_AVAILABLE:
        return "(AI not available)"
    return client.chat.completions.create(model=os.environ.get("OPENAI_MODEL", "google/gemini-2.5-flash"), messages=[{"role": "user", "content": prompt}], temperature=0.3).choices[0].message.content

np.random.seed(42)
genes = ["BRCA1", "TP53", "EGFR", "MYC", "KRAS", "BRAF", "PIK3CA", "PTEN", "RB1", "APC"]
control = np.random.poisson(lam=100, size=(len(genes), 3))
treatment = np.random.poisson(lam=100, size=(len(genes), 3))
treatment[0] *= 3; treatment[1] *= 4; treatment[4] //= 3; treatment[7] //= 2

df = pd.DataFrame(np.hstack([control, treatment]), index=genes, columns=["c1","c2","c3","t1","t2","t3"])
df["ctrl_mean"] = df[["c1","c2","c3"]].mean(axis=1)
df["treat_mean"] = df[["t1","t2","t3"]].mean(axis=1)
df["log2_fc"] = np.log2((df["treat_mean"] / df["ctrl_mean"]).replace(0, np.nan))

print("Differential expression results:")
for gene in df.sort_values("log2_fc", ascending=False).index:
    fc = df.loc[gene, "log2_fc"]
    direction = "UP" if fc > 1 else ("DOWN" if fc < -1 else "---")
    print(f"  {gene:8s} log2FC={fc:+.2f}  {direction}")

up_genes = df[df["log2_fc"] > 1].index.tolist()
down_genes = df[df["log2_fc"] < -1].index.tolist()

print(f"\n--- AI: What do these expression changes mean? ---\n")
result = ask_ai(
    f"In a cancer treatment experiment, these genes changed expression:\n\n"
    f"Upregulated: {up_genes}\n"
    f"Downregulated: {down_genes}\n\n"
    "Please explain:\n"
    "1. What does each gene do? (BRCA1, TP53, KRAS, PTEN)\n"
    "2. Is it good or bad that TP53 is upregulated in a cancer context?\n"
    "3. What pathway might be affected?\n"
    "4. What experiment would you do next to confirm these results?\n\n"
    "Explain for a college student who knows basic biology but not cancer biology."
)
print(result)

#!/usr/bin/env python3


import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

np.random.seed(42)
n_genes = 200

# Simulate log2 fold changes and p-values
# Most genes: no change
log2fc = np.random.normal(0, 0.5, n_genes)
# Some genes: real changes
log2fc[0:10] = np.random.uniform(1.5, 4, 10)    # upregulated
log2fc[10:20] = np.random.uniform(-4, -1.5, 10)  # downregulated

# P-values: significant for changed genes, random for others
neg_log_p = np.random.exponential(0.5, n_genes)
neg_log_p[0:10] = np.random.uniform(3, 10, 10)   # significant
neg_log_p[10:20] = np.random.uniform(3, 10, 10)  # significant

# Classify
colors = []
for i in range(n_genes):
    if abs(log2fc[i]) > 1 and neg_log_p[i] > 1.3:  # 1.3 = -log10(0.05)
        colors.append("red" if log2fc[i] > 0 else "blue")
    else:
        colors.append("gray")

# Plot
fig, ax = plt.subplots(figsize=(8, 6))
ax.scatter(log2fc, neg_log_p, c=colors, alpha=0.6, s=20)
ax.axhline(y=1.3, color='black', linestyle='--', alpha=0.3, label='p=0.05')
ax.axvline(x=-1, color='black', linestyle='--', alpha=0.3)
ax.axvline(x=1, color='black', linestyle='--', alpha=0.3)
ax.set_xlabel('log2(Fold Change)')
ax.set_ylabel('-log10(p-value)')
ax.set_title('Volcano Plot: Differential Expression')
ax.legend(['p=0.05', 'Upregulated', 'Downregulated', 'Not significant'],
          loc='upper right')

outfile = '/tmp/volcano.png'
plt.savefig(outfile, dpi=150, bbox_inches='tight')
print(f"Volcano plot saved to {outfile}")
print(f"Total genes: {n_genes}")
print(f"Upregulated (red): {sum(1 for c in colors if c == 'red')}")
print(f"Downregulated (blue): {sum(1 for c in colors if c == 'blue')}")
print(f"Not significant (gray): {sum(1 for c in colors if c == 'gray')}")

#!/usr/bin/env python3


import os
import numpy as np

try:
    from openai import OpenAI
    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", ""), base_url=os.environ.get("OPENAI_BASE_URL", "https://openrouter.ai/api/v1"))
    AI_AVAILABLE = True
except ImportError:
    AI_AVAILABLE = False
    print("Note: Install openai package for AI features\n")

def ask_ai(prompt: str) -> str:
    if not AI_AVAILABLE:
        return "(AI not available)"
    return client.chat.completions.create(model=os.environ.get("OPENAI_MODEL", "google/gemini-2.5-flash"), messages=[{"role": "user", "content": prompt}], temperature=0.3).choices[0].message.content

# Simulate results
np.random.seed(42)
n = 200
log2fc = np.random.normal(0, 0.5, n)
log2fc[0:10] = np.random.uniform(1.5, 4, 10)
log2fc[10:20] = np.random.uniform(-4, -1.5, 10)
pval = np.random.exponential(0.5, n)
pval[0:20] = np.random.uniform(3, 10, 20)

up = sum(1 for i in range(n) if log2fc[i] > 1 and pval[i] > 1.3)
down = sum(1 for i in range(n) if log2fc[i] < -1 and pval[i] > 1.3)

print(f"Simulation: {n} genes, {up} upregulated, {down} downregulated")

print("\n--- AI: How do I read a volcano plot? ---\n")
result = ask_ai(
    f"I made a volcano plot from RNA-seq data:\n"
    f"- {n} genes tested\n"
    f"- {up} significantly upregulated (log2FC > 1, p < 0.05)\n"
    f"- {down} significantly downregulated (log2FC < -1, p < 0.05)\n\n"
    "Please explain:\n"
    "1. What does each axis of a volcano plot represent?\n"
    "2. Why is it called a 'volcano' plot?\n"
    "3. What do the genes in the top-right corner mean biologically?\n"
    "4. How do scientists decide the cutoffs for 'significant'?\n"
    "5. What is the difference between statistical significance and biological significance?\n\n"
    "Use analogies for a college freshman."
)
print(result)

#!/usr/bin/env python3


# Simulated GO annotations for our differentially expressed genes
go_annotations = {
    "BRCA1": {
        "biological_process": ["DNA repair", "cell cycle checkpoint", "double-strand break repair"],
        "molecular_function": ["ubiquitin-protein ligase activity", "protein binding"],
        "cellular_component": ["nucleus", "BRCA1-A complex"]
    },
    "TP53": {
        "biological_process": ["apoptosis", "cell cycle arrest", "DNA damage response"],
        "molecular_function": ["DNA binding", "transcription factor activity"],
        "cellular_component": ["nucleus", "cytoplasm"]
    },
    "KRAS": {
        "biological_process": ["signal transduction", "cell proliferation", "MAPK cascade"],
        "molecular_function": ["GTPase activity", "protein binding"],
        "cellular_component": ["plasma membrane", "cytoplasm"]
    },
    "PTEN": {
        "biological_process": ["negative regulation of cell proliferation", "apoptosis"],
        "molecular_function": ["phosphatase activity", "phosphoprotein phosphatase activity"],
        "cellular_component": ["cytoplasm", "nucleus"]
    }
}

print("Gene Ontology annotations for key genes:")
print("=" * 60)

for gene, go in go_annotations.items():
    print(f"\n{gene}:")
    for aspect, terms in go.items():
        label = aspect.replace("_", " ").title()
        print(f"  {label}:")
        for term in terms:
            print(f"    - {term}")

# Find shared GO terms
print("\n\nShared biological processes:")
bp_sets = {}
for gene, go in go_annotations.items():
    for term in go["biological_process"]:
        bp_sets.setdefault(term, []).append(gene)

for term, genes in sorted(bp_sets.items(), key=lambda x: -len(x[1])):
    if len(genes) > 1:
        print(f"  {term}: {', '.join(genes)}")

#!/usr/bin/env python3


import os

try:
    from openai import OpenAI
    client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", ""), base_url=os.environ.get("OPENAI_BASE_URL", "https://openrouter.ai/api/v1"))
    AI_AVAILABLE = True
except ImportError:
    AI_AVAILABLE = False
    print("Note: Install openai package for AI features\n")

def ask_ai(prompt: str) -> str:
    if not AI_AVAILABLE:
        return "(AI not available)"
    return client.chat.completions.create(model=os.environ.get("OPENAI_MODEL", "google/gemini-2.5-flash"), messages=[{"role": "user", "content": prompt}], temperature=0.3).choices[0].message.content

genes = ["BRCA1", "TP53", "KRAS", "PTEN"]
processes = ["DNA repair", "apoptosis", "signal transduction", "cell proliferation"]

print("Key differentially expressed genes and their functions:")
for g, p in zip(genes, processes):
    print(f"  {g}: {p}")

print("\n--- AI: What pathway connects these genes? ---\n")
result = ask_ai(
    f"In a cancer experiment, these genes were differentially expressed:\n"
    f"- BRCA1 (DNA repair) -- upregulated\n"
    f"- TP53 (apoptosis) -- upregulated\n"
    f"- KRAS (signal transduction) -- downregulated\n"
    f"- PTEN (tumor suppressor) -- downregulated\n\n"
    "Please explain:\n"
    "1. What biological pathway connects these genes?\n"
    "2. Are these changes consistent with a treatment working or failing?\n"
    "3. Which of these genes are known drug targets?\n"
    "4. What is the PI3K/AKT pathway and how does PTEN regulate it?\n\n"
    "Draw a simple pathway diagram using text arrows."
)
print(result)