Skip to content

Correlation heatmap

Example

import gwaslab as gl
import pandas as pd

Example

gl.show_version()

stdout: 

2024/12/23 12:20:37 GWASLab v4.0.0 https://cloufield.github.io/gwaslab/
2024/12/23 12:20:37 (C) 2022-2026, Yunye He, Kamatani Lab, GPL-3.0 license, gwaslab@gmail.com

Load sample data

Sample data source:

  • Kanai, M., Akiyama, M., Takahashi, A., Matoba, N., Momozawa, Y., Ikeda, M., ... & Kamatani, Y. (2018). Genetic analysis of quantitative traits in the Japanese population links cell types to complex human diseases. Nature genetics, 50(3), 390-400.

Example

ldsc = pd.read_csv("../0_sample_data/toy_data/input_rg.txt",sep="\t")
ldsc
p1_category p1 p2_category p2 rg se z p q
Anthropometric Height Anthropometric BMI -0.0587 0.0240 -2.4421 0.014602 0.136798
Anthropometric Height Metabolic TC -0.0778 0.0344 -2.2634 0.023611 0.188695
Anthropometric Height Metabolic HDL-C -0.0045 0.0364 -0.1230 0.902080 0.971590
Anthropometric Height Metabolic LDL-C -0.1245 0.0426 -2.9228 0.003469 0.050126
Anthropometric Height Metabolic TG -0.0426 0.0309 -1.3792 0.167820 0.523652
... ... ... ... ... ... ... ... ...
Other Pollinosis Tumor PrCa 0.3250 0.1433 2.2683 0.023300 0.187944
Other Pollinosis Tumor UF 0.2163 0.1756 1.2315 0.218100 0.596616
Other Pollinosis Other Urolithiasis 0.0664 0.1794 0.3701 0.711300 0.903018
Other Urolithiasis Allergic disease AD -0.0821 0.1592 -0.5159 0.605890 0.863730
Other Urolithiasis Tumor UF 0.1980 0.1292 1.5324 0.125410 0.456418

[3916 rows x 9 columns]

Full heatmap

Example

df = gl.plot_rg( ldsc,
            p="q",
            p1="p2",
            p2="p1", 
            sig_levels=[0.05],
            corrections =["non"],
            full_cell=("non",0.05),
            panno_texts=["*"],
            panno_kwargs={"size":12,"c":"black"},
            fig_kwargs={"figsize":(15,15),"dpi":300},
            colorbar_kwargs={"shrink":0.4},
            fontsize=8
            )

stdout: 

2024/12/23 12:20:37 Start to create ldsc genetic correlation heatmap...
2024/12/23 12:20:37 Raw dataset records: 3916
2024/12/23 12:20:37  -Raw dataset non-NA records: 3916
2024/12/23 12:20:37 Filling diagonal line and duplicated pair for plotting...
2024/12/23 12:20:37  -Diagonal records: 89
2024/12/23 12:20:37 Valid unique trait pairs: 3916
2024/12/23 12:20:37  -Valid unique trait1: 88
2024/12/23 12:20:37  -Valid unique trait2: 88
2024/12/23 12:20:37  -Significant correlations with P < 0.05: 270
2024/12/23 12:20:37  -Significant correlations after Bonferroni correction: 81
2024/12/23 12:20:37  -Significant correlations with FDR <0.05: 127
2024/12/23 12:20:37 Plotting heatmap...
2024/12/23 12:20:37 Full cell : non-corrected P == 0.05
2024/12/23 12:20:42 P value annotation text (Order: Bon -> FDR -> Pnom): 
2024/12/23 12:20:42  -* : non-corrected P < 0.05 
2024/12/23 12:20:42 Start to save figure...
2024/12/23 12:20:42  -Skip saving figure!
2024/12/23 12:20:42 Finished saving figure...
2024/12/23 12:20:42 Finished creating ldsc genetic correlation heatmap!

Output image

filter and order the data

Example

# Order the data
trait =  pd.read_csv("../0_sample_data/toy_data/trait_list.txt",sep="\t")
trait["order"] = range(len(trait))
trait

# These steps are just used to order the traits

order = trait["TRAIT"].values

trait_set1 = trait.loc[trait["order"]>=59,"TRAIT"].values
trait_set2 = trait.loc[trait["order"]<59,"TRAIT"].values

ldsc = ldsc.loc[((ldsc["p1"].isin(trait_set1))&(ldsc["p2"].isin(trait_set2))) | ((ldsc["p1"].isin(trait_set2))&(ldsc["p2"].isin(trait_set1))),:]

# key for sort
map_dic={order[i]:i+1 for i in range(len(order))}
key=lambda x:x.map(map_dic)

Example

print(map_dic)

stdout: 

{'Height': 1, 'BMI': 2, 'TC': 3, 'HDL-C': 4, 'LDL-C': 5, 'TG': 6, 'BS': 7, 'HbA1c': 8, 'TP': 9, 'Alb': 10, 'NAP': 11, 'A/G': 12, 'BUN': 13, 'sCr': 14, 'eGFR': 15, 'UA': 16, 'Na': 17, 'K': 18, 'Cl': 19, 'Ca': 20, 'P': 21, 'TBil': 22, 'ZTT': 23, 'AST': 24, 'ALT': 25, 'ALP': 26, 'GGT': 27, 'APTT': 28, 'PT': 29, 'Fbg': 30, 'CK': 31, 'LDH': 32, 'CRP': 33, 'WBC': 34, 'Neutro': 35, 'Eosino': 36, 'Baso': 37, 'Mono': 38, 'Lym': 39, 'RBC': 40, 'Hb': 41, 'Ht': 42, 'MCV': 43, 'MCH': 44, 'MCHC': 45, 'Plt': 46, 'SBP': 47, 'DBP': 48, 'MAP': 49, 'PP': 50, 'IVS': 51, 'PW': 52, 'LVDd': 53, 'LVDs': 54, 'LVM': 55, 'LVMI': 56, 'RWT': 57, 'FS': 58, 'EF': 59, 'T2D': 60, 'IS': 61, 'CeAn': 62, 'MI': 63, 'PAD': 64, 'AF': 65, 'Asthma': 66, 'AD': 67, 'GD': 68, 'RA': 69, 'CHB': 70, 'CHC': 71, 'Anemia': 72, 'BD': 73, 'SCZ': 74, 'AIS': 75, 'Osteoporosis': 76, 'LuCa': 77, 'GaCa': 78, 'EsCa': 79, 'CoCa': 80, 'PrCa': 81, 'BrCa': 82, 'EnCa': 83, 'UF': 84, 'Glaucoma': 85, 'COPD': 86, 'Epilepsy': 87, 'Pollinosis': 88, 'Urolithiasis': 89}

Replicate heatmap in paper

Example

df = gl.plot_rg( ldsc,
            sig_levels=[0.05],
            corrections =["non"],
            p="q",
            p1="p2",
            p2="p1",
            full_cell=("non",0.05),
            panno_texts=["*"],
            fig_kwargs={"figsize":(15,15),"dpi":300},
            colorbar_kwargs={"shrink":0.4},
            panno_kwargs={"size":12,"c":"black"},
            fontsize=8,
            sort_key=key
            )

stdout: 

2024/12/23 12:20:43 Start to create ldsc genetic correlation heatmap...
2024/12/23 12:20:43 Raw dataset records: 1770
2024/12/23 12:20:43  -Raw dataset non-NA records: 1770
2024/12/23 12:20:43 Filling diagonal line and duplicated pair for plotting...
2024/12/23 12:20:43 Valid unique trait pairs: 1770
2024/12/23 12:20:43  -Valid unique trait1: 59
2024/12/23 12:20:43  -Valid unique trait2: 30
2024/12/23 12:20:43  -Significant correlations with P < 0.05: 68
2024/12/23 12:20:43  -Significant correlations after Bonferroni correction: 13
2024/12/23 12:20:43  -Significant correlations with FDR <0.05: 20
2024/12/23 12:20:43 Plotting heatmap...
2024/12/23 12:20:43 Full cell : non-corrected P == 0.05
2024/12/23 12:20:44 P value annotation text (Order: Bon -> FDR -> Pnom): 
2024/12/23 12:20:44  -* : non-corrected P < 0.05 
2024/12/23 12:20:44 Start to save figure...
2024/12/23 12:20:44  -Skip saving figure!
2024/12/23 12:20:44 Finished saving figure...
2024/12/23 12:20:44 Finished creating ldsc genetic correlation heatmap!

Output image