🔮 PLASMA

PLASMA (PopuLation Allele-Specific MApping) is a statistical fine-mapping method for functional data using QTL and allelic-imbalance signal.

Preprint for the PLASMA method

Developed at the Gusev Lab at the Dana Farber Cancer Institute / Harvard Medical School.

Installation and Dependencies

PLASMA utilizes Python 2.7 requires the following Python packages for core functionality:

• Numpy 1.15+
• Scipy 1.1+

The following packages are optional, but are used for pre/post-processing:

• Pandas 0.23+
• Matplotlib 2.2+
• Seaborn 0.9+
• Pysam 0.15+
• PyVCF 0.6+

All packages can be installed using Python's pip package manager.

To download PLASMA, click "Clone or Download" or enter:

git clone https://github.com/austintwang/plasma

run_plasma.py : Quick-start fine-mapping script

The run_plasma.py script conducts fine-mapping of a single locus with default PLASMA parameters and outputs.

Input Files and Parameters

The script requires the following files:

• Two text files (one for each haplotype), specifying the haplotype-specific genotypes, across samples and marker. Each row should represent an individual, and each column should represent a marker. The ordering of samples and markers should be the same for both files.
• Two text files (one for each haplotype), specifying the haplotype-specific phenotypes (e.g. read counts) across samples. The order of samples should be the same as that of the genotype files.
• A text file, specifying the total phenotype across samples. If none is provided, then the total phenotype is assumed to be the sum of the haplotype-specific phenotypes.

Other parameters include:

• Individual-level or global beta-binomial overdispersions.
• AS-Only and QTL-only modes, where the total phenotype and allele-specific phenotypes are ignored, respectively.
• Search parameters, including the maximum number of causal variants and the search mode (exhaustive or stochastic shotgun search)
• The confidence level when creating the credible set

Output Files

The script outputs two files in the specified output directory:

• cset.txt: The minimal set of markers that contains the set of true causal markers, at the specified confidence level. 1 and 0 denote that a marker is included in and excluded from the credible set, respectively. The order of the markers is the same as that in the genotype files.
• ppas.txt: The marginal posterior probabilities of each marker being causal.

Usage

Usage of the script is as follows:

usage: run_plasma.py [-h] [--total_exp_path TOTAL_EXP_PATH]
                     [--overdispersion_path OVERDISPERSION_PATH]
                     [--overdispersion_global OVERDISPERSION_GLOBAL]
                     [--as_only] [--qtl_only] [--search_mode SEARCH_MODE]
                     [--max_causal MAX_CAUSAL] [--confidence CONFIDENCE]
                     hap_A_path counts_A_path hap_B_path counts_B_path out_dir

positional arguments:
  hap_A_path            Path to haplotype A genotypes file
  counts_A_path         Path to haplotype A mapped counts file
  hap_B_path            Path to haplotype B genotypes file
  counts_B_path         Path to haplotype B mapped counts file
  out_dir               Path to output directory

optional arguments:
  -h, --help            show this help message and exit
  --total_exp_path TOTAL_EXP_PATH, -t TOTAL_EXP_PATH
                        Path to total QTL phenotype file (Default: Sum of
                        counts files)
  --overdispersion_path OVERDISPERSION_PATH, -o OVERDISPERSION_PATH
                        Path to individual-level AS overdispersion file
                        (Default: Global overdispersion)
  --overdispersion_global OVERDISPERSION_GLOBAL, -g OVERDISPERSION_GLOBAL
                        Global AS overdispersion (Default: 0)
  --as_only, -a         AS-Only Mode
  --qtl_only, -q        QTL-Only Mode
  --search_mode SEARCH_MODE, -s SEARCH_MODE
                        Causal configuration search mode (Default:
                        "exhaustive")
  --max_causal MAX_CAUSAL, -m MAX_CAUSAL
                        Maximum number of causal configurations searched
                        (Default: 1)
  --confidence CONFIDENCE, -c CONFIDENCE
                        Credible set confidence level (Default: 0.95)

PLASMA API

PLASMA additionally has a Python API, which exposes the full feature set of PLASMA. Documentation for the PLASMA Python API is currently in progress.

Features of the API include:

• Alternative data input formats, including direct use of association statistics
• User specification of hyperparameters, including heritabilities and correlations between the AS and QTL phenotypes
• Additional fine-mapping outputs
• Colocalization analysis across multiple quantitative allele-specific phenotypes
• An allele-specific simulation framework for quantitative phenotypes
• Ability to extend PLASMA via subtyping

To see the latest code, check the dev branch.