beast_visualiser.py

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

# **📌 BEAST log dosyasını oku**
log_file = "/home/analysis/Desktop/re_root/new_dawn/_1739865905/new_analysis2_BEAST.log"

# **BEAST log dosyasını DataFrame'e çevir**
df = pd.read_csv(log_file, sep="\t", comment="#")

# **Gerekli sütunları al**
df_filtered = df[["dN", "dS", "dN/dS", "posterior"]].dropna()

# **Ka/Ks oranına göre seçilim sınıflandırması**
df_filtered["Selection Type"] = df_filtered["dN/dS"].apply(lambda x: 
    "Positive Selection" if x > 1 else ("Neutral" if 0.9 <= x <= 1.1 else "Negative Selection")
)

# **Ortalama Ka/Ks hesapla**
mean_dnds = df_filtered["dN/dS"].mean()

# **Bayes faktörü ile dN/dS arasındaki ilişkiyi görselleştir**
plt.figure(figsize=(10, 6))
sns.scatterplot(x=df_filtered["posterior"], y=df_filtered["dN/dS"], hue=df_filtered["Selection Type"], palette="coolwarm", alpha=0.7)
plt.axhline(1, color="red", linestyle="--", label="Neutral Evolution (Ka/Ks = 1)")
plt.xlabel("Bayesian Posterior Probability")
plt.ylabel("Ka/Ks (dN/dS) Ratio")
plt.title("Bayesian Support vs Ka/Ks (dN/dS) Ratio")
plt.legend()
plt.show()

# **Histogram: Ka/Ks Dağılımı**
plt.figure(figsize=(10, 6))
sns.histplot(df_filtered["dN/dS"], bins=20, kde=True, color="blue")
plt.axvline(1, color="red", linestyle="--", label="Neutral Evolution (Ka/Ks = 1)")
plt.xlabel("Ka/Ks (dN/dS) Ratio")
plt.ylabel("Frequency")
plt.title(f"Ka/Ks (dN/dS) Distribution - Mean: {mean_dnds:.2f}")
plt.legend()
plt.show()

# **Boxplot ile Seçilim Tiplerini Göster**
plt.figure(figsize=(8, 6))
sns.boxplot(x=df_filtered["Selection Type"], y=df_filtered["dN/dS"], palette="coolwarm")
plt.axhline(1, color="red", linestyle="--", label="Neutral Evolution (Ka/Ks = 1)")
plt.ylabel("Ka/Ks Ratio")
plt.xlabel("Selection Type")
plt.title("Selection Pressure on Genes")
plt.legend()
plt.show()

# **Sonuçları göster**
import ace_tools as tools
tools.display_dataframe_to_user(name="Advanced BEAST dN/dS Analysis", dataframe=df_filtered)