### Install weasyprint for PDF Generation

Source: https://cloufield.github.io/gwaslab/QC_Report

Provides instructions for installing the `weasyprint` library, which is a dependency for generating PDF reports. It includes both pip installation and system-level package installation for Ubuntu/Debian.

```bash
pip install weasyprint

# Ubuntu/Debian:
sudo apt-get install python3-weasyprint

```

--------------------------------

### Install bcftools

Source: https://cloufield.github.io/gwaslab/AssignCHRPOS

Instructions for installing the `bcftools` utility, which is a prerequisite for both `.rsid_to_chrpos2()` and `gl.process_vcf_to_hfd5()` functions in GWASLab for processing VCF files.

```bash
# Using conda:
conda install -c bioconda bcftools

# Or download from the official website:
# https://www.htslib.org/download/

# Verify installation:
bcftools --version
```

--------------------------------

### Get GWASLab Help and Version

Source: https://cloufield.github.io/gwaslab/CLI

Displays the help message for GWASLab, providing information on available commands and options. Also shows the installed version of the GWASLab tool.

```bash
# Show help message
gwaslab --help

# Show version
gwaslab version
```

--------------------------------

### Install GWASLab Package

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Installs the GWASLab package version 4.0.0 using pip. This command ensures that the library is available for use in the current environment.

```shell
# !pip install gwaslab==4.0.0
```

--------------------------------

### Show GWASLab CLI Version

Source: https://cloufield.github.io/gwaslab/CLI

Displays the installed version of the GWASLab CLI. This is a simple command to verify the installation and check for updates.

```bash
gwaslab version
```

--------------------------------

### Install GWASLab using uv

Source: https://cloufield.github.io/gwaslab/index

Installs the latest version of GWASLab using uv, a fast Python package installer written in Rust. This method also requires a compatible Python version.

```bash
uv pip install -U gwaslab

```

--------------------------------

### Install and Import GWASLab from GitHub

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Clones the GWASLab repository from GitHub and adds its source directory to the Python system path. This allows using the latest development version of the library.

```python
#!git clone https://github.com/Cloufield/gwaslab.git

import sys
sys.path.insert(0,"/home/yunye/work/gwaslab/src")
import gwaslab as gl
```

--------------------------------

### Comprehensive QC Report Generation Example

Source: https://cloufield.github.io/gwaslab/QC_Report

A complete example demonstrating the generation of a comprehensive QC report in PDF format with customized options for basic QC, lead variant extraction, MQQ plots, and regional plots. It also includes setting a custom report title.

```python
import gwaslab as gl

# Load sumstats
mysumstats = gl.Sumstats(
    "examples/0_sample_data/t2d_bbj.txt.gz",
    snpid="SNP",
    chrom="CHR",
    pos="POS",
    ea="ALT",
    nea="REF",
    neaf="Frq",
    beta="BETA",
    se="SE",
    p="P",
    build="19",
    n="N"
)

# Generate comprehensive QC report with regional plots
report_path = mysumstats.report(
    output_path="t2d_qc_report.pdf",
    report_title="T2D GWAS QC Report",
    basic_check_kwargs={
        "remove": False,
        "remove_dup": False,
        "normalize": True,
        "verbose": False
    },
    get_lead_kwargs={
        "sig_level": 5e-8,
        "windowsizekb": 500,
        "anno": False,
        "verbose": False
    },
    mqq_plot_kwargs={
        "save": True,
        "dpi": 300,
        "verbose": False
    },
    regional_plot_kwargs={
        "vcf_path": "/path/to/reference.vcf.gz",
        "region_size": 500,
        "save": True,
        "dpi": 300,
        "verbose": False
    }
)

print(f"Report generated: {report_path}")

```

--------------------------------

### Complete Pipeline with All Options in GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Executes a comprehensive pipeline including QC, harmonization, and format conversion with numerous options using the GWASLab CLI. This example showcases advanced filtering, reference file usage, and output formatting for LDSC. Requires input sumstats and various reference files.

```bash
# Complete pipeline: QC + Harmonization + Format conversion
gwaslab --input raw_sumstats.tsv \
  --fmt auto \
  --qc \
  --remove \
  --remove-dup \
  --normalize \
  --harmonize \
  --ref-seq /data/reference/hg19.fasta.gz \
  --ref-rsid-vcf /data/reference/dbsnp.vcf.gz \
  --ref-infer /data/reference/1kg.vcf.gz \
  --ref-alt-freq AF \
  --maf-threshold 0.40 \
  --ref-maf-threshold 0.4 \
  --sweep-mode \
  --threads 8 \
  --out final_sumstats \
  --to-fmt ldsc \
  --tab-fmt tsv \
  --exclude-hla \
  --hapmap3 \
  --n 100000 \
  --chr-prefix chr \
  --xymt-number

```

--------------------------------

### Import GWASLab and Pandas Libraries

Source: https://cloufield.github.io/gwaslab/visualization_brisbane

Initializes the environment by importing the GWASLab library as 'gl' and the pandas library as 'pd'. These are essential for subsequent operations in the script.

```python
import gwaslab as gl
import pandas as pd
```

--------------------------------

### Download Sample Data using wget

Source: https://cloufield.github.io/gwaslab/visualization_brisbane

Downloads an Excel file containing genetic association data from a specified URL using the 'wget' command. This data is required for the subsequent plotting and analysis steps.

```shell
!wget https://static-content.springer.com/esm/art%3A10.1038%2Fs41586-022-05275-y/MediaObjects/41586_2022_5275_MOESM3_ESM.xlsx
```

--------------------------------

### BED File Format Example

Source: https://cloufield.github.io/gwaslab/QC%26Filtering

Illustrates the standard format for a BED (Browser Extensible Data) file used for defining genomic regions. Each line specifies a chromosome, start position, and end position.

```text
chr1    1000000    2000000
chr2    5000000    6000000
chr3    10000000   11000000
```

--------------------------------

### Example Verbose Output for Ancestry Inference

Source: https://cloufield.github.io/gwaslab/InferAncestry

This example illustrates the detailed logging output produced by `mysumstats.infer_ancestry` when `verbose=True`. It shows the Fst values calculated for different populations, the variant count used, and the final inferred closest ancestry. This output is crucial for understanding the inference process and potential data quality issues.

```text
Start to infer ancestry based on Fst ...(version)
 -Estimating Fst using 12345 variants...
 -FST_GBR : 0.001234
 -FST_FIN : 0.001456
 -FST_CHS : 0.002345
 ...
 -FST_EUR : 0.001234
 -Closest Ancestry: EUR
Finished inferring ancestry.
```

--------------------------------

### SSF Validation Output Examples

Source: https://cloufield.github.io/gwaslab/Format

Provides examples of the output messages from the SSF validation process in GWASLab. It shows both successful validation and specific error messages when validation fails, such as missing required columns.

```text
✓ SSF validation successful: File passes all validation checks


```

```text
✗ SSF validation failed: Missing required columns: ['standard_error']
  - Missing required columns: ['standard_error']
```

--------------------------------

### Download Sample Data using wget

Source: https://cloufield.github.io/gwaslab/visualization_regional

Downloads a sample summary statistics file (t2d_bbj.txt.gz) using the wget command-line utility. This file is used in subsequent examples.

```bash
#!wget -O t2d_bbj.txt.gz http://jenger.riken.jp/14/
```

--------------------------------

### Load Summary Statistics with GWASLab

Source: https://cloufield.github.io/gwaslab/harmonization_workflow

This example demonstrates how to load summary statistics from a TSV file using the `gl.Sumstats` class. It specifies the file path, the format ('gwaslab'), and additional columns to include ('NOTE'). The output shows the loading process and initial data verification.

```python
mysumstats = gl.Sumstats("/home/yunye/work/gwaslab/examples/toy_data/to_harmonize.tsv",fmt="gwaslab",other=["NOTE"])
```

--------------------------------

### Multi-step Processing with GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Demonstrates a multi-step processing workflow using the GWASLab CLI, including QC, harmonization, and final format conversion. Each step takes the output of the previous step as input, allowing for modular data processing.

```bash
# Step 1: QC
gwaslab --input raw.tsv --fmt auto --qc --remove-dup --out step1_qc --to-fmt gwaslab

# Step 2: Harmonize
gwaslab --input step1_qc.gwaslab.tsv.gz --fmt gwaslab --harmonize \
  --ref-seq ref.fasta --out step2_harmonized --to-fmt gwaslab

# Step 3: Format for specific tool
gwaslab --input step2_harmonized.gwaslab.tsv.gz --fmt gwaslab \
  --out final --to-fmt plink --tab-fmt tsv

```

--------------------------------

### Download and Load GWAS Data

Source: https://cloufield.github.io/gwaslab/EffectSize

Downloads example GWAS summary statistics files for male and female BMI from a remote URL and loads them into Sumstats objects. This sets up the data for subsequent analysis.

```python
!wget -O bbj_bmi_male.txt.gz http://jenger.riken.jp/2analysisresult_qtl_download/
!wget -O bbj_bmi_female.txt.gz http://jenger.riken.jp/4analysisresult_qtl_download/

# Load sumstats as Sumstats objects
gl1 = gl.Sumstats("bbj_bmi_female.txt.gz",fmt="gwaslab",snpid="SNP",ea="REF",nea="ALT",sep="\t")
gl2 = gl.Sumstats("bbj_bmi_male.txt.gz",fmt="gwaslab",snpid="SNP",ea="REF",nea="ALT",sep="\t")
```

--------------------------------

### Output with bgzip and tabix Indexing using GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Creates a bgzip-compressed file with a tabix index for the output sumstats using the GWASLab CLI. This is useful for creating indexed files that can be efficiently queried. Requires input sumstats.

```bash
# Create bgzip-compressed file with tabix index
gwaslab --input sumstats.tsv \
  --fmt gwaslab \
  --out indexed_sumstats \
  --to-fmt gwaslab \
  --bgzip \
  --tabix

```

--------------------------------

### GWASLab CLI with Filtering Options

Source: https://cloufield.github.io/gwaslab/Format

Demonstrates advanced CLI usage of GWASLab, including options for filtering data. This example shows how to apply HapMap3 filtering, exclude HLA variants, and limit the number of variants to 10,000.

```bash
gwaslab --input sumstats.tsv --fmt gwaslab --out output --to-fmt gwaslab \
  --hapmap3 --exclude-hla --n 10000
```

--------------------------------

### Example Usage of Regional Plot with LD Link

Source: https://cloufield.github.io/gwaslab/RegionalPlot

Provides a Python code example demonstrating how to use the `plot_mqq` function for regional plotting with LD link visualization enabled. It shows how to specify the plot mode, region, VCF path for LD reference, and parameters for LD link, including alpha scaling, line width, and significance level.

```python
mysumstats.plot_mqq(
    mode="r",
    region=(7, 156938803, 157038803),
    vcf_path="path/to/reference.vcf.gz",
    ld_link=True,
    ld_link_alpha_scale=0.2,
    ld_link_linewidth=1.0,
    ld_link_sig_level=5e-8  # Only show lines for significant variants
)
```

--------------------------------

### Display GWASLab Version and Environment Information

Source: https://cloufield.github.io/gwaslab/harmonization_liftover

Calls the 'show_version' function from the GWASLab library to display the installed version, license information, and the Python environment details. This is useful for debugging and ensuring compatibility.

```python
gl.show_version()
```

--------------------------------

### Download Sample Data

Source: https://cloufield.github.io/gwaslab/visualization_miami2

These commands use wget to download sample GWAS summary statistics files. The files are saved with specific names and paths, intended for use in subsequent analysis or examples.

```bash
#!wget -O ../0_sample_data/bmi_male_bbj.txt.gz http://jenger.riken.jp/2analysisresult_qtl_download/
#!wget -O ../0_sample_data/bmi_female_bbj.txt.gz http://jenger.riken.jp/4analysisresult_qtl_download/

```

--------------------------------

### Gene Annotation Workflow Example

Source: https://cloufield.github.io/gwaslab/GTF

Illustrates a common gene annotation workflow: reading GTF with essential columns, filtering for protein-coding genes, and then extracting gene coordinates for display.

```python
from gwaslab.io.io_gtf import read_gtf

# Read genes with essential information
gt = read_gtf(
    "annotation.gtf.gz",
    features={"gene"},
    usecols=["seqname", "start", "end", "strand", "gene_id", "gene_name", "gene_biotype"]
)

# Filter for protein-coding genes
protein_coding = gt[gt["gene_biotype"] == "protein_coding"]

# Get gene coordinates
for _, gene in protein_coding.iterrows():
    print(f"{gene['gene_name']}: {gene['seqname']}:{gene['start']}-{gene['end']}")

```

--------------------------------

### Generate HTML and PDF QC Reports

Source: https://cloufield.github.io/gwaslab/QC_Report

Demonstrates how to generate QC reports in both HTML and PDF formats by changing the file extension in the `output_path`. PDF generation requires the `weasyprint` library to be installed.

```python
# Generate HTML report
mysumstats.report("report.html")

# Generate PDF report (requires weasyprint)
mysumstats.report("report.pdf")

```

--------------------------------

### Format Conversion for LDSC with GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Converts sumstats to LDSC format for heritability analysis using the GWASLab CLI. It takes an input sumstats file and specifies the output format as 'ldsc', with an option to disable gzip compression.

```bash
# Convert to LDSC format for heritability analysis
gwaslab --input sumstats.tsv \
  --fmt gwaslab \
  --out sumstats_ldsc \
  --to-fmt ldsc \
  --no-gzip

```

--------------------------------

### List Files in Sample Data Directory

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Lists the files present in the cloned gwaslab-sample-data directory. This helps verify that the sample data has been downloaded correctly.

```shell
!ls  gwaslab-sample-data
```

--------------------------------

### Get Number to NCBI Conversion Table - gwaslab Python

Source: https://cloufield.github.io/gwaslab/harmonization_workflow

Retrieves a dictionary mapping chromosome numbers to NCBI Sequence Identifiers for a specified build. This is useful for standardizing chromosome names in VCF files.

```python
gl.get_number_to_NC(build="19")
```

--------------------------------

### Clone GWASLab Sample Data Repository

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Clones the gwaslab-sample-data repository from GitHub to download necessary sample files for the tutorial. This is a prerequisite for loading and processing data.

```shell
#!git clone https://github.com/Cloufield/gwaslab-sample-data.git
```

--------------------------------

### Assign RSIDs with VCF to BCF Conversion

Source: https://cloufield.github.io/gwaslab/AssignrsID

This example demonstrates assigning RSIDs from a VCF file after converting it to BCF format for improved performance. It also includes options to `strip_info` fields to reduce file size. Requires `bcftools` to be installed.

```python
# Convert VCF to BCF first for better performance
mysumstats.basic_check()
mysumstats.assign_rsid2(
    vcf_path="/path/to/dbsnp/GCF_000001405.25.vcf.gz",
    convert_to_bcf=True,  # Convert to BCF format
    strip_info=True,       # Strip INFO fields to reduce size
    threads=8
)
```

--------------------------------

### Basic Format Conversion Examples

Source: https://cloufield.github.io/gwaslab/Format

Demonstrates basic usage of the .to_format() function for converting summary statistics to different formats like LDSC and PLINK. It shows how to specify the output path, format, tabular format, and whether to gzip the output.

```python
# Convert to LDSC format
mysumstats.to_format(path="./output", fmt="ldsc")
# Output: output.ldsc.tsv.gz

# Convert to PLINK format with CSV
mysumstats.to_format(path="./output", fmt="plink", tab_fmt="csv", gzip=False)
# Output: output.plink.csv

# Convert to VCF with bgzip and tabix index
mysumstats.to_format(path="./output", fmt="vcf", bgzip=True, tabix=True)
# Output: output.vcf.bcf.gz and output.vcf.bcf.gz.tbi
```

--------------------------------

### Verify bcftools Installation

Source: https://cloufield.github.io/gwaslab/AssignrsID

This command verifies the installation and version of bcftools, which is a prerequisite for the sweep mode (.assign_rsid2()) in GWASLab. Successful execution shows the bcftools version and its associated htslib version.

```bash
bcftools --version

```

--------------------------------

### Install GWASLab in a conda environment

Source: https://cloufield.github.io/gwaslab/index

Sets up a conda environment with Python 3.12 and installs GWASLab using pip. Alternatively, it shows how to create an environment from a yml file.

```bash
conda create -n gwaslab -c conda-forge python=3.12

conda activate gwaslab

pip install -U gwaslab

```

```bash
conda env create -n gwaslab -f environment.yml -c conda-forge

```

--------------------------------

### Install GWASLab using pip

Source: https://cloufield.github.io/gwaslab/index

Installs the latest version of GWASLab using pip. This method requires Python 3.9 or later, with Python 3.12 recommended for optimal performance and compatibility.

```bash
pip install -U gwaslab

```

--------------------------------

### Prepare Data for Meta-analysis with GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Prepares data for meta-analysis by performing QC, harmonization, and formatting using the GWASLab CLI. It includes options for removing duplicates, excluding the HLA region, and specifying sample size (N). Requires input sumstats and reference files.

```bash
# QC, harmonize, and format for meta-analysis
gwaslab --input study1.tsv \
  --fmt auto \
  --qc \
  --remove-dup \
  --harmonize \
  --ref-seq /data/reference/hg19.fasta.gz \
  --ref-rsid-vcf /data/reference/dbsnp.vcf.gz \
  --out study1_ready \
  --to-fmt gwaslab \
  --exclude-hla \
  --n 50000

```

--------------------------------

### Example Usage of GWASLab Clumping Function

Source: https://cloufield.github.io/gwaslab/Clumping

This example demonstrates how to use the `clump` function from the `mysumstats` module in GWASLab. It shows a basic call to the function, specifying a VCF file as input for the clumping process.

```python
clumps = mysumstats.clump(vcf= "/home/yunye/.gwaslab/EAS.ALL.split_norm_af.1kgp3v5.hg19.vcf.gz")

```

--------------------------------

### Initialize and Display Sumstats Data

Source: https://cloufield.github.io/gwaslab/utility_data_conversion

This example shows how to initialize a `Sumstats` object with data from a gzipped text file, specifying column mappings for SNP ID, chromosome, position, beta, and standard error. It also demonstrates how to display the first few rows of the loaded data and then modifies the 'SE' column to simulate extreme P-values.

```python
mysumstats = gl.Sumstats("../0_sample_data/t2d_bbj.txt.gz",
             snpid="SNP",
             chrom="CHR",
             pos="POS",
             beta="BETA",
             se="SE",nrows=5, verbose=False)
# simulate some extreme P values by shrinking the SE
mysumstats.data["SE"] = mysumstats.data["SE"]/100
print(mysumstats.data)
```

--------------------------------

### Customize MQQ Plot Appearance

Source: https://cloufield.github.io/gwaslab/visualization_brisbane

Customizes the appearance of the MQQ plot. This example enables annotation, sets the significance line color to red, increases the window size for density calculation, and suppresses verbose output. It utilizes the 'b' mode for the Brisbane plot.

```python
sumstats.plot_mqq(mode="b",anno=True,sig_line_color="red", windowsizekb=100000, verbose=False)
```

--------------------------------

### Download Reference Data with gl.download_ref

Source: https://cloufield.github.io/gwaslab/tutorial_v4

This function downloads reference genome data, such as GTF files for specific builds. It requires the name of the reference dataset to be downloaded. No specific input validation or output is shown, but it's essential for downstream analyses requiring genomic annotations.

```python
gl.download_ref("ensembl_hg19_gtf")
```

--------------------------------

### Download Reference Data for GWASLab

Source: https://cloufield.github.io/gwaslab/AssignrsID

Downloads necessary reference data for rsID assignment. This is a one-time setup. Supported references include '1kg_dbsnp151_hg19_auto' and '1kg_dbsnp151_hg38_auto'.

```python
# 1. Download reference data (one-time setup)
gl.download_ref("1kg_dbsnp151_hg19_auto")
```

--------------------------------

### Download Reference Files using GWASLab

Source: https://cloufield.github.io/gwaslab/tutorial_v4

This function allows users to download preprocessed reference datasets for use in GWAS analysis. The downloaded files are stored in the ~/.gwaslab directory by default. Ensure sufficient disk space for downloads, such as the ~2.8GB for '1kg_eas_hg19'.

```python
# gl.download_ref("1kg_eas_hg19")
```

--------------------------------

### Method Chaining for Report Generation (Python)

Source: https://cloufield.github.io/gwaslab/QC_Report

Demonstrates how to use the `report()` method with method chaining to generate a GWAS report. This example shows initializing a `Sumstats` object, performing basic checks, identifying lead variants, and then generating an HTML report.

```python
mysumstats = gl.Sumstats("data.txt.gz", ...)

result = (mysumstats
    .basic_check(remove=False, normalize=True)
    .get_lead(sig_level=5e-8)
    .report("report.html")
)
```

--------------------------------

### Get Reference GTF Data

Source: https://cloufield.github.io/gwaslab/GTF

Retrieves reference GTF data for a specified build and source, such as Ensembl hg38. It demonstrates filtering the retrieved data for protein-coding genes.

```python
from gwaslab.io.io_gtf import get_gtf

# Get Ensembl hg38 GTF for chromosome 1
gt = get_gtf(chrom="1", build="38", source="ensembl")

# Filter for protein-coding genes
protein_coding = gt[gt["gene_biotype"] == "protein_coding"]
print(f"Protein-coding genes on chr1: {len(protein_coding)}")

```

--------------------------------

### Create QQ Plot with GWASLab

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Generates a Quantile-Quantile (QQ) plot from GWAS summary statistics. It processes variants, calculates lambda GC, and annotates the plot with gene names. The plot can be saved to a file.

```python
mysumstats.plot_mqq(output="my_first_mqq_plot.png", GENENAME="GENENAME", repel_force=0.03)
```

--------------------------------

### Basic QC Pipeline with GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Loads, performs quality control, and saves sumstats using the GWASLab CLI. It handles input file loading, auto-detection of format, QC steps (ID, chromosome, position, allele fixing), removal of low-quality variants and duplicates, indel normalization, and output saving. Requires input sumstats file.

```bash
# Load, QC, and save sumstats
gwaslab --input raw_sumstats.tsv \
  --fmt auto \
  --qc \
  --remove \
  --remove-dup \
  --normalize \
  --threads 4 \
  --out cleaned_sumstats \
  --to-fmt gwaslab \
  --quiet

```

--------------------------------

### Testing with Small Sample using GWASLab CLI

Source: https://cloufield.github.io/gwaslab/CLIWorkflowExamples

Tests the GWASLab CLI command on the first 1000 rows of a large sumstats file. This is useful for quick checks and debugging before running on the full dataset. Includes QC and duplicate removal options.

```bash
# Test command on first 1000 rows
gwaslab --input large_sumstats.tsv \
  --fmt auto \
  --nrows 1000 \
  --qc \
  --remove-dup \
  --out test_output \
  --to-fmt gwaslab

```

--------------------------------

### Get Chromosome List - GWASLab

Source: https://cloufield.github.io/gwaslab/CommonData

Retrieves a list of chromosomes for a specified species, with options to include numeric representations or only numeric autosomes. Supports species-specific chromosome naming conventions.

```python
gl.get_chr_list()
# ['1','2','3',...,'22','X','Y','MT']

gl.get_chr_list(species="mouse")
# ['1','2','3',...,'19','X','Y','MT']

gl.get_chr_list(species="chicken")
# ['1','2','3',...,'28','Z','W','MT']

gl.get_chr_list(add_number=True)
# ['1','2',...,'22','X','Y','MT',1,2,...,22]

gl.get_chr_list(only_number=True)
# [1, 2, 3, ..., 22]
```

--------------------------------

### Display GWASLab Version

Source: https://cloufield.github.io/gwaslab/visualization_miami2

This command displays the installed version of the gwaslab library and its associated license and contact information. It's a simple utility function to verify the library's version.

```python
import gwaslab as gl

gl.show_version()

```

--------------------------------

### Load Sumstats by Specifying Columns

Source: https://cloufield.github.io/gwaslab/SumstatsObject

Demonstrates how to load summary statistics by manually specifying the column names for each required and optional field.

```APIDOC
## Load sumstats by specifying columns

### Description
This example shows how to load GWAS summary statistics from a file by manually mapping the column names to the expected fields in the Sumstats Object.

### Method
`gl.Sumstats(file_path, snpid='column_name', chrom='column_name', pos='column_name', ea='column_name', nea='column_name', eaf='column_name', beta='column_name', se='column_name', p='column_name', n='column_name', ...)`

### Parameters
(See `gl.Sumstats()` constructor for a full list of parameters and their descriptions.)

### Request Example
```python
mysumstats = gl.Sumstats("t2d_bbj.txt.gz",
             snpid="SNPID",
             chrom="CHR",
             pos="POS",
             ea="Allele2",
             nea="Allele1",
             eaf="AF_Allele2",
             beta="BETA",
             se="SE",
             p="p.value",
             n="N")
```

### Response
#### Success Response (200)
- **Sumstats Object**: A GWASLab Sumstats Object populated with data from the specified file and column mappings.
```

--------------------------------

### Load Summary Statistics into GWASLab Sumstats Object

Source: https://cloufield.github.io/gwaslab/tutorial_v4

Loads summary statistics from a gzipped text file into a GWASLab Sumstats object. It requires specifying column names for various genetic association parameters and the file's separator.

```python
mysumstats = gl.Sumstats("gwaslab-sample-data/bbj_t2d_hm3_chr7_variants.txt.gz",
             snpid="SNPID",
             chrom="CHR",
             pos="POS",
             ea="EA",
             nea="NEA",
             neaf="EAF",
             beta="BETA",
             se="SE",
             p="P",
             n="N",
             sep="\t")
```

--------------------------------

### Testing with Small Samples (GWASLab CLI)

Source: https://cloufield.github.io/gwaslab/CLI

Demonstrates how to test commands on a subset of data using the `--nrows` option. This is helpful for debugging and verifying command functionality before running on the entire dataset.

```bash
gwaslab --input large_sumstats.tsv --fmt auto --nrows 1000 --qc --out test_output --to-fmt gwaslab
```

--------------------------------

### Input Format Detection (GWASLab CLI)

Source: https://cloufield.github.io/gwaslab/CLI

Demonstrates the use of `--fmt auto` to automatically detect the input file format. This simplifies the process by removing the need to manually specify the input format, making it easier to process various sumstats files.

```bash
gwaslab --input sumstats.tsv --fmt auto --qc --out output --to-fmt gwaslab
```

--------------------------------

### Performance Tips for GTF Handling

Source: https://cloufield.github.io/gwaslab/GTF

Provides recommendations for optimizing GTF file processing performance within GWASLab.

```APIDOC
## Performance Tips

1.  **Filter early** : Use `chrom` and `features` parameters to filter data during reading, not after loading.
2.  **Select columns** : Use `usecols` to load only the columns you need, reducing memory usage.
3.  **Use built-in references** : For common reference genomes, use `get_gtf()` which automatically handles downloading and caching.
4.  **Compressed files** : GWASLab automatically handles gzip-compressed GTF files, so you can work directly with `.gtf.gz` files.
5.  **Polars backend** : The GTF reader uses Polars internally for fast file reading, then converts to pandas for compatibility.
```

--------------------------------

### Get Default Download Directory with GWASLab

Source: https://cloufield.github.io/gwaslab/Download

Retrieves the currently set default directory for downloaded files. This function returns the path as a string, allowing users to confirm their download location settings.

```python
gl.get_default_directory()
```

--------------------------------

### Infer Genome Build - Output Logs - Python

Source: https://cloufield.github.io/gwaslab/InferBuild

Example logging output from the .infer_build() function, showing the process of matching variants and inferring the genome build. It also includes a warning for unreliable inference due to a limited number of matching variants.

```python
 -Loading Hapmap3 variants data...
 -CHR and POS will be used for matching...
 -Matching variants for hg19: num_hg19 = 12345
 -Matching variants for hg38: 5678
 -Since num_hg19 >> num_hg38, set the genome build to hg19 for the STATUS code....

⚠️ Please be cautious due to the limited number of variants.
```

--------------------------------

### Common Filtering Workflow Example

Source: https://cloufield.github.io/gwaslab/QC%26Filtering

Demonstrates a typical sequence of filtering steps for genetic association data, including removing palindromic variants, indels, excluding the HLA region, and retaining HapMap3 variants for LDSC analysis.

```python
# Remove palindromic variants
mysumstats.filter_palindromic(mode="out")

# Remove indels
mysumstats.filter_indel(mode="out")

# Exclude HLA region
mysumstats.exclude_hla()

# Filter to HapMap3 variants (for LDSC)
mysumstats.filter_hapmap3()
```

--------------------------------

### Run Basic QC Workflow in gwaslab

Source: https://cloufield.github.io/gwaslab/Standardization

Demonstrates how to load GWAS summary statistics and run the basic quality control workflow using the `basic_check()` method in the gwaslab library. It shows default usage, removal of bad variants, and customization of individual steps.

```python
import gwaslab as gl

# Load sumstats
mysumstats = gl.Sumstats("sumstats.txt.gz", fmt="plink2")

# Run comprehensive QC (doesn't remove variants by default)
mysumstats.basic_check()

# Run QC and remove bad variants
mysumstats.basic_check(remove=True, remove_dup=True)

# Customize individual steps
mysumstats.basic_check(
    remove=True,
    remove_dup=True,
    normalize=True,
    threads=4,
    fix_chr_kwargs={"remove": True},
    fix_pos_kwargs={"limit": 300000000},
    sanity_check_stats_kwargs={"p": (1e-300, 1)}
)

```

--------------------------------

### Save Annotated Effect Size Comparison Plot

Source: https://cloufield.github.io/gwaslab/EffectSize

Performs an annotated comparison of effect sizes between two GWAS datasets and saves the resulting plot to a file. This example includes options for plot customization and saving.

```python
# GWASLab will automatically extract significant variants from both sumstats. 
a = gl.compare_effect(path1=gl1,
                      path2=gl2,
                      label=["Female","Male","Both","None"],
                      xylabel_prefix="Per-allele effect size for ",
                      anno=True,
                      anno_het=True,
                      anno_diff=0.015,
                      sig_level=5e-8,
                      legend_title=r'$mathregular{ P < 5 x 10^{-8}}$ in:',
                      verbose=True,
                      save="myplot.png",
                      save_kwargs={"dpi":300,"facecolor":"white"}
)
```