biopipen 0.25.4__py3-none-any.whl → 0.26.1__py3-none-any.whl

biopipen/__init__.py

@@ -1 +1 @@
-__version__ = "0.25.4"
+__version__ = "0.26.1"

biopipen/core/config.toml

@@ -27,6 +27,8 @@ convert = "convert"
 wget = "wget"
 # aria2c
 aria2c = "aria2c"
+# plink
+plink = "plink"
 # tabix
 tabix = "tabix"
 # sambamba

biopipen/ns/rnaseq.py

@@ -5,17 +5,154 @@ from ..core.config import config
 
 
 class UnitConversion(Proc):
-    """Convert expression value units back and forth"""
+    """Convert expression value units back and forth
+
+    See <https://haroldpimentel.wordpress.com/2014/05/08/what-the-fpkm-a-review-rna-seq-expression-units/>
+    and <https://docs.gdc.cancer.gov/Data/Bioinformatics_Pipelines/Expression_mRNA_Pipeline/#fpkm>.
+
+    Following converstions are supported -
+    * `count -> cpm, fpkm/rpkm, fpkmuq/rpkmrq, tpm, tmm`
+    * `fpkm/rpkm -> count, tpm, cpm`
+    * `tpm -> count, fpkm/rpkm, cpm`
+    * `cpm -> count, fpkm/rpkm, tpm`
+    NOTE that during some conversions, `sum(counts/effLen)` is approximated to
+    `sum(counts)/sum(effLen) * length(effLen))`
+
+    You can also use this process to just transform the expression values, e.g., take
+    log2 of the expression values. In this case, you can set `inunit` and `outunit` to
+    `count` and `log2(count + 1)` respectively.
+
+    Input:
+        infile: Input file containing expression values
+            The file should be a matrix with rows representing genes and columns
+            representing samples.
+            It could be an RDS file containing a data frame or a matrix, or a
+            text file containing a matrix with tab as the delimiter. The text
+            file can be gzipped.
+
+    Output:
+        outfile: Output file containing the converted expression values
+            The file will be a matrix with rows representing genes and columns
+            representing samples.
+
+    Envs:
+        inunit: The input unit of the expression values.
+            You can also use an expression to indicate the input unit, e.g.,
+            `log2(counts + 1)`. The expression should be like `A * fn(B*X + C) + D`,
+            where `A`, `B`, `C` and `D` are constants, `fn` is a function, and X is
+            the input unit.
+            Currently only `expr`, `sqrt`, `log2`, `log10` and `log` are supported as
+            functions.
+            Supported input units are:
+            * counts/count/rawcounts/rawcount: raw counts.
+            * cpm: counts per million.
+            * fpkm/rpkm: fragments per kilobase of transcript per million.
+            * fpkmuq/rpkmuq: upper quartile normalized FPKM/RPKM.
+            * tpm: transcripts per million.
+            * tmm: trimmed mean of M-values.
+        outunit: The output unit of the expression values. An expression can also be
+            used for transformation (e.g. `log2(tpm + 1)`). If `inunit` is `count`,
+            then this means we are converting raw counts to tpm, and transforming it
+            to `log2(tpm + 1)` as the output. Any expression supported by `R` can be
+            used. Same units as `inunit` are supported.
+        refexon: Path to the reference exon gff file.
+        meanfl (type=auto): A file containing the mean fragment length for each sample
+            by rows (samples as rowname), without header.
+            Or a fixed universal estimated number (1 used by TCGA).
+        nreads (type=auto): The estimatied total number of reads for each sample.
+            or you can pass a file with the number for each sample by rows
+            (samples as rowname), without header.
+            When converting `fpkm/rpkm -> count`, it should be total reads of that sample.
+            When converting `cpm -> count`: it should be total reads of that sample.
+            When converting `tpm -> count`: it should be total reads of that sample.
+            When converting `tpm -> cpm`: it should be total reads of that sample.
+            When converting `tpm -> fpkm/rpkm`: it should be `sum(fpkm)` of that sample.
+            It is not used when converting `count -> cpm, fpkm/rpkm, tpm`.
+    """  # noqa: E501
     input = "infile:file"
     output = "outfile:file:{{in.infile | basename}}"
     lang = config.lang.rscript
     envs = {
-        "infmt": "matrix",  # or rds
         "inunit": None,
         "outunit": None,
         "refexon": config.ref.refexon,
-        "meanfl": None,
-        "inlog2p": False,
-        "outlog2p": False,
+        "meanfl": 1,
+        "nreads": 1_000_000,
     }
     script = "file://../scripts/rnaseq/UnitConversion.R"
+
+
+class Simulation(Proc):
+    """Simulate RNA-seq data using ESCO/RUVcorr package
+
+    Input:
+        ngenes: Number of genes to simulate
+        nsamples: Number of samples to simulate
+            If you want to force the process to re-simulate for the same
+            `ngenes` and `nsamples`, you can set a different value for `envs.seed`.
+            Note that the samples will be shown as cells in the output (since
+            the simulation is designed for single-cell RNA-seq data).
+
+    Output:
+        outfile: Output file containing the simulated data with rows representing
+            genes and columns representing samples.
+        outdir: Output directory containing the simulated data
+            `sim.rds` and `True.rds` will be generated.
+            For `ESCO`, `sim.rds` contains the simulated data in a
+            `SingleCellExperiment` object, and `True.rds` contains the matrix of true
+            counts.
+            For `RUVcorr`, `sim.rds` contains the simulated data in list with
+            `Truth`, A matrix containing the values of Xβ; `Y` A matrix containing the
+            values in `Y`; `Noise` A matrix containing the values in `Wα`; `Sigma`
+            A matrix containing the true gene-gene correlations, as defined by Xβ; and
+            `Info` A matrix containing some of the general information about the
+            simulation.
+            For all matrices, rows represent genes and columns represent samples.
+
+    Envs:
+        tool (choice): Which tool to use for simulation.
+            - ESCO: uses the [ESCO](https://github.com/JINJINT/ESCO) package.
+            - RUVcorr: uses the [RUVcorr](https://rdrr.io/bioc/RUVcorr/) package.
+        ncores (type=int): Number of cores to use.
+        seed (type=int): Random seed.
+            If not set, seed will not be set.
+        esco_args (ns): Additional arguments to pass to the simulation function.
+            - save (choice): Which type of data to save to `out.outfile`.
+                - `simulated-truth`: saves the simulated true counts.
+                - `zero-inflated`: saves the zero-inflated counts.
+                - `down-sampled`: saves the down-sampled counts.
+            - type (choice): Which type of heterogenounity to use.
+                - single: produces a single population.
+                - group: produces distinct groups.
+                - tree: produces distinct groups but admits a tree structure.
+                - traj: produces distinct groups but admits a smooth trajectory
+                    structure.
+            - <more>: See <https://rdrr.io/github/JINJINT/ESCO/man/escoParams.html>.
+        ruvcorr_args (ns): Additional arguments to pass to the simulation
+            function.
+            - <more>: See <https://rdrr.io/bioc/RUVcorr/man/simulateGEdata.html>.
+        transpose_output (flag): If set, the output will be transposed.
+        index_start (type=int): The index to start from when naming the samples.
+            Affects the sample names in `out.outfile` only.
+    """
+    input = "ngenes:var, nsamples:var"
+    output = [
+        "outfile:file:{{in.ngenes}}x{{in.nsamples}}.sim/simulated.txt",
+        "outdir:dir:{{in.ngenes}}x{{in.nsamples}}.sim",
+    ]
+    lang = config.lang.rscript
+    envs = {
+        "tool": "RUVcorr",
+        "ncores": config.misc.ncores,
+        "type": "single",
+        "esco_args": {
+            "dropout-type": "none",
+            "save": "simulated-truth",
+            "type": "single",
+        },
+        "ruvcorr_args": {},
+        "seed": None,
+        "transpose_output": False,
+        "index_start": 1,
+    }
+    script = "file://../scripts/rnaseq/Simulation.R"

biopipen/ns/scrna.py

@@ -483,14 +483,18 @@ class SeuratClusterStats(Proc):
             The parameters from the cases can overwrite the default parameters.
             - frac (flag): Whether to output the fraction of cells instead of number.
             - pie (flag): Also output a pie chart?
+            - circos (flag): Also output a circos plot?
             - table (flag): Whether to output a table (in tab-delimited format) and in the report.
             - frac_ofall(flag): Whether to output the fraction against all cells,
                 instead of the fraction in each group.
+                Does not work for circos plot.
                 Only works when `frac` is `True` and `group-by` is specified.
             - transpose (flag): Whether to transpose the cluster and group, that is,
                 using group as the x-axis and cluster to fill the plot.
+                For circos plot, when transposed, the arrows will be drawn from the idents (by `ident`) to the
+                the groups (by `group-by`).
                 Only works when `group-by` is specified.
-            - position (choice): The position of the bars.
+            - position (choice): The position of the bars. Does not work for pie and circos plots.
                 - stack: Use `position_stack()`.
                 - fill: Use `position_fill()`.
                 - dodge: Use `position_dodge()`.
@@ -499,8 +503,13 @@ class SeuratClusterStats(Proc):
             - group-by: The column name in metadata to group the cells.
                 Does NOT support for pie charts.
             - split-by: The column name in metadata to split the cells into different plots.
+                Does NOT support for circos plots.
             - subset: An expression to subset the cells, will be passed to
                 `dplyr::filter()` on metadata.
+            - circos_devpars (ns): The device parameters for the circos plots.
+                - res (type=int): The resolution of the plots.
+                - height (type=int): The height of the plots.
+                - width (type=int): The width of the plots.
             - pie_devpars (ns): The device parameters for the pie charts.
                 - res (type=int): The resolution of the plots.
                 - height (type=int): The height of the plots.
@@ -634,6 +643,7 @@ class SeuratClusterStats(Proc):
         "stats_defaults": {
             "frac": False,
             "pie": False,
+            "circos": False,
             "table": False,
             "position": "auto",
             "frac_ofall": False,
@@ -644,6 +654,7 @@ class SeuratClusterStats(Proc):
             "subset": None,
             "devpars": {"res": 100, "height": 600, "width": 800},
             "pie_devpars": {"res": 100, "height": 600, "width": 800},
+            "circos_devpars": {"res": 100, "height": 600, "width": 600},
         },
         "stats": {
             "Number of cells in each cluster": {
@@ -882,8 +893,9 @@ class CellsDistribution(Proc):
         each: The column name in metadata to separate the cells into different plots.
         section: The section to show in the report. This allows different cases to be put in the same section in report.
             Only works when `each` is not specified.
-        overlap (list): Plot the overlap of cells in different cases under the same section.
-            The section must have at least 2 cases.
+        overlap (list): Plot the overlap of cell groups (values of `cells_by`) in different cases
+            under the same section.
+            The section must have at least 2 cases, each case should have a single `cells_by` column.
         cases (type=json;order=99): If you have multiple cases, you can specify them here.
             Keys are the names of the cases and values are the options above except `mutaters`.
             If some options are not specified, the options in `envs` will be used.
@@ -1141,6 +1153,7 @@ class TopExpressingGenes(Proc):
             markers See below for all libraries.
             <https://maayanlab.cloud/Enrichr/#libraries>
         n (type=int): The number of top expressing genes to find.
+        subset: An expression to subset the cells for each case.
         cases (type=json): If you have multiple cases, you can specify them
             here. The keys are the names of the cases and the values are the
             above options except `mutaters`. If some options are
@@ -1161,6 +1174,7 @@ class TopExpressingGenes(Proc):
         "section": "DEFAULT",
         "dbs": ["KEGG_2021_Human", "MSigDB_Hallmark_2020"],
         "n": 250,
+        "subset": None,
         "cases": {},
     }
     plugin_opts = {

biopipen/ns/snp.py

@@ -0,0 +1,70 @@
+"""Plink processes"""
+
+from ..core.proc import Proc
+from ..core.config import config
+
+
+class PlinkSimulation(Proc):
+    """Simulate SNPs using PLINK v1.9
+
+    See also <https://www.cog-genomics.org/plink/1.9/input#simulate>.
+
+    Input:
+        nsnps: Number of SNPs to simulate
+        ncases: Number of cases to simulate
+        nctrls: Number of controls to simulate
+
+    Output:
+        outdir: Output directory containing the simulated data
+            `plink_sim.bed`, `plink_sim.bim`, and `plink_sim.fam` will be generated.
+        gtmat: Genotype matrix file containing the simulated data with rows representing
+            SNPs and columns representing samples.
+
+    Envs:
+        plink: Path to PLINK v1.9
+        seed (type=int): Random seed.
+            If not set, seed will not be set.
+        label: Prefix label for the SNPs.
+        prevalence  (type=float): Disease prevalence.
+        minfreq (type=float): Minimum allele frequency.
+        maxfreq (type=float): Maximum allele frequency.
+        hetodds (type=float): Odds ratio for heterozygous genotypes.
+        homodds (type=float): Odds ratio for homozygous genotypes.
+        missing (type=float): Proportion of missing genotypes.
+        args (ns): Additional arguments to pass to PLINK.
+            - <more>: see <https://www.cog-genomics.org/plink/1.9/input#simulate>.
+        transpose_gtmat (flag): If set, the genotype matrix (`out.gtmat`) will
+            be transposed.
+        sample_prefix: Use this prefix for the sample names. If not set, the sample
+            names will be `per0_per0`, `per1_per1`, `per2_per2`, etc. If set, the
+            sample names will be `prefix0`, `prefix1`, `prefix2`, etc.
+            This only affects the sample names in the genotype matrix file
+            (`out.gtmat`).
+    """
+    input = "nsnps:var, ncases:var, nctrls:var"
+    output = [
+        (
+            "outdir:dir:{{in.nsnps | int}}_"
+            "{{in.ncases | int}}xcases_{{in.nctrls | int}}xctrls.plink_sim"
+        ),
+        (
+            "gtmat:file:{{in.nsnps | int}}_"
+            "{{in.ncases | int}}xcases_{{in.nctrls | int}}xctrls.plink_sim/gtmat.txt"
+        ),
+    ]
+    lang = config.lang.python
+    envs = {
+        "plink": config.exe.plink,
+        "seed": None,
+        "label": "SNP",
+        "prevalence": 0.01,
+        "minfreq": 0.0,
+        "maxfreq": 1.0,
+        "hetodds": 1.0,
+        "homodds": 1.0,
+        "missing": 0.0,
+        "args": {},
+        "transpose_gtmat": False,
+        "sample_prefix": None,
+    }
+    script = "file://../scripts/snp/PlinkSimulation.py"

biopipen/ns/stats.py

@@ -0,0 +1,320 @@
+"""Provides processes for statistics."""
+
+from ..core.proc import Proc
+from ..core.config import config
+
+
+class ChowTest(Proc):
+    """Massive Chow tests.
+
+    See Also https://en.wikipedia.org/wiki/Chow_test
+
+    Input:
+        infile: The input data file. The rows are samples and the columns are
+            features. It must be tab-delimited.
+            ```
+            Sample   F1   F2   F3   ...   Fn
+            S1       1.2  3.4  5.6        7.8
+            S2       2.3  4.5  6.7        8.9
+            ...
+            Sm       5.6  7.8  9.0        1.2
+            ```
+        groupfile: The group file. The rows are the samples and the columns
+            are the groupings. It must be tab-delimited.
+            ```
+            Sample   G1   G2   G3   ...   Gk
+            S1       0    1    0          0
+            S2       2    1    0          NA  # exclude this sample
+            ...
+            Sm       1    0    0          0
+            ```
+        fmlfile: The formula file. The first column is grouping and the
+            second column is the formula. It must be tab-delimited.
+            ```
+            Group   Formula  ...  # Other columns to be added to outfile
+            G1      Fn ~ F1 + Fx + Fy   # Fx, Fy could be covariates
+            G1      Fn ~ F2 + Fx + Fy
+            ...
+            Gk      Fn ~ F3 + Fx + Fy
+            ```
+
+    Output:
+        outfile: The output file. It is a tab-delimited file with the first
+            column as the grouping and the second column as the p-value.
+            ```
+            Group  Formula  ...  Pooled  Groups  SSR  SumSSR  Fstat  Pval  Padj
+            G1     Fn ~ F1       0.123   2       1    0.123   0.123  0.123 0.123
+            G1     Fn ~ F2       0.123   2       1    0.123   0.123  0.123 0.123
+            ...
+            Gk     Fn ~ F3       0.123   2       1    0.123   0.123  0.123 0.123
+            ```
+
+    Envs:
+        padj (choice): The method for p-value adjustment.
+            - none: No p-value adjustment (no Padj column in outfile).
+            - holm: Holm-Bonferroni method.
+            - hochberg: Hochberg method.
+            - hommel: Hommel method.
+            - bonferroni: Bonferroni method.
+            - BH: Benjamini-Hochberg method.
+            - BY: Benjamini-Yekutieli method.
+            - fdr: FDR correction method.
+        transpose_input (flag): Whether to transpose the input file.
+        transpose_group (flag): Whether to transpose the group file.
+    """
+    input = "infile:file, groupfile:file, fmlfile:file"
+    output = "outfile:file:{{in.infile | stem}}.chowtest.txt"
+    lang = config.lang.rscript
+    envs = {
+        "padj": "none",
+        "transpose_input": False,
+        "transpose_group": False,
+    }
+    script = "file://../scripts/stats/ChowTest.R"
+
+
+class LiquidAssoc(Proc):
+    """Liquid association tests.
+
+    See Also https://github.com/gundt/fastLiquidAssociation
+    Requieres https://github.com/pwwang/fastLiquidAssociation
+
+    Input:
+        infile: The input data file. The rows are samples and the columns are
+            features. It must be tab-delimited.
+            ```
+            Sample   F1   F2   F3   ...   Fn
+            S1       1.2  3.4  5.6        7.8
+            S2       2.3  4.5  6.7        8.9
+            ...
+            Sm       5.6  7.8  9.0        1.2
+            ```
+            The features (columns) will be tested pairwise, which will be the X and
+            Y columns in the result of `fastMLA`
+        covfile: The covariate file. The rows are the samples and the columns
+            are the covariates. It must be tab-delimited.
+            If provided, the data in `in.infile` will be adjusted by covariates by
+            regressing out the covariates and the residuals will be used for
+            liquid association tests.
+        groupfile: The group file. The rows are the samples and the columns
+            are the groupings. It must be tab-delimited.
+            ```
+            Sample   G1   G2   G3   ...   Gk
+            S1       0    1    0          0
+            S2       2    1    0          NA  # exclude this sample
+            ...
+            Sm       1    0    0          0
+            ```
+            This will be served as the Z column in the result of `fastMLA`
+            This can be omitted. If so, `envs.nvec` should be specified, which is
+            to select column from `in.infile` as Z.
+        fmlfile: The formula file. The 3 columns are X3, X12 and X21. The results
+            will be filtered based on the formula. It must be tab-delimited without
+            header.
+
+    Output:
+        outfile: The output file.
+            ```
+            X12  X21  X3  rhodiff  MLA value  estimates  san.se  wald  Pval  model
+            C38  C46  C5  0.87  0.32  0.67  0.20  10.87  0  F
+            C46  C38  C5  0.87  0.32  0.67  0.20  10.87  0  F
+            C27  C39  C4  0.94  0.34  1.22  0.38  10.03  0  F
+            ```
+
+    Envs:
+        nvec: The column index (1-based) of Z in `in.infile`, if `in.groupfile` is
+            omitted. You can specify multiple columns by comma-seperated values, or
+            a range of columns by `-`. For example, `1,3,5-7,9`. It also supports
+            column names. For example, `F1,F3`. `-` is not supported for column
+            names.
+        x: Similar as `nvec`, but limit X group to given features.
+            The rest of features (other than X and Z) in `in.infile` will
+            be used as Y.
+            The features in `in.infile` will still be tested pairwise, but only
+            features in X and Y will be kept.
+        topn (type=int): Number of results to return by `fastMLA`, ordered from
+            highest `|MLA|` value descending.
+            The default of the package is 2000, but here we set to 1e6 to return as
+            many results as possible (also good to do pvalue adjustment).
+        rvalue (type=float): Tolerance value for LA approximation. Lower values of
+            rvalue will cause a more thorough search, but take longer.
+        cut (type=int): Value passed to the GLA function to create buckets
+            (equal to number of buckets+1). Values placing between 15-30 samples per
+            bucket are optimal. Must be a positive integer>1. By default,
+            `max(ceiling(nrow(data)/22), 4)` is used.
+        ncores (type=int): Number of cores to use for parallelization.
+        padj (choice): The method for p-value adjustment.
+            - none: No p-value adjustment (no Padj column in outfile).
+            - holm: Holm-Bonferroni method.
+            - hochberg: Hochberg method.
+            - hommel: Hommel method.
+            - bonferroni: Bonferroni method.
+            - BH: Benjamini-Hochberg method.
+            - BY: Benjamini-Yekutieli method.
+            - fdr: FDR correction method.
+        transpose_input (flag): Whether to transpose the input file.
+        transpose_group (flag): Whether to transpose the group file.
+        transpose_cov (flag): Whether to transpose the covariate file.
+        xyz_names: The names of X12, X21 and X3 in the final output file. Separated
+            by comma. For example, `X12,X21,X3`.
+    """
+    input = "infile:file, covfile:file, groupfile:file, fmlfile:file"
+    output = "outfile:file:{{in.infile | stem}}.liquidassoc.txt"
+    lang = config.lang.rscript
+    envs = {
+        "nvec": None,
+        "x": None,
+        "topn": 1e6,
+        "rvalue": 0.5,
+        "cut": 20,
+        "ncores": config.misc.ncores,
+        "padj": "none",
+        "transpose_input": False,
+        "transpose_group": False,
+        "transpose_cov": False,
+        "xyz_names": None,
+    }
+    script = "file://../scripts/stats/LiquidAssoc.R"
+
+
+class DiffCoexpr(Proc):
+    """Differential co-expression analysis.
+
+    See also <https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-11-497>
+    and <https://github.com/DavisLaboratory/dcanr/blob/8958d61788937eef3b7e2b4118651cbd7af7469d/R/inference_methods.R#L199>.
+
+    Input:
+        infile: The input data file. The rows are samples and the columns are
+            features. It must be tab-delimited.
+            ```
+            Sample   F1   F2   F3   ...   Fn
+            S1       1.2  3.4  5.6        7.8
+            S2       2.3  4.5  6.7        8.9
+            ...
+            Sm       5.6  7.8  9.0        1.2
+            ```
+        groupfile: The group file. The rows are the samples and the columns
+            are the groupings. It must be tab-delimited.
+            ```
+            Sample   G1   G2   G3   ...   Gk
+            S1       0    1    0          0
+            S2       2    1    0          NA  # exclude this sample
+            ...
+            Sm       1    0    0          0
+            ```
+
+    Output:
+        outfile: The output file. It is a tab-delimited file with the first
+            column as the feature pair and the second column as the p-value.
+            ```
+            Group  Feature1  Feature2  Pval  Padj
+            G1     F1        F2        0.123 0.123
+            G1     F1        F3        0.123 0.123
+            ...
+            ```
+
+    Envs:
+        method (choice): The method used to calculate the differential
+            co-expression.
+            - pearson: Pearson correlation.
+            - spearman: Spearman correlation.
+        beta: The beta value for the differential co-expression analysis.
+        padj (choice): The method for p-value adjustment.
+            - none: No p-value adjustment (no Padj column in outfile).
+            - holm: Holm-Bonferroni method.
+            - hochberg: Hochberg method.
+            - hommel: Hommel method.
+            - bonferroni: Bonferroni method.
+            - BH: Benjamini-Hochberg method.
+            - BY: Benjamini-Yekutieli method.
+            - fdr: FDR correction method.
+        perm_batch (type=int): The number of permutations to run in each batch
+        seed (type=int): The seed for random number generation
+        ncores (type=int): The number of cores to use for parallelization
+        transpose_input (flag): Whether to transpose the input file.
+        transpose_group (flag): Whether to transpose the group file.
+    """  # noqa: E501
+    input = "infile:file, groupfile:file"
+    output = "outfile:file:{{in.infile | stem}}.diffcoexpr.txt"
+    lang = config.lang.rscript
+    envs = {
+        "method": "pearson",
+        "beta": 6,
+        "padj": "none",
+        "perm_batch": 20,
+        "seed": 8525,
+        "ncores": config.misc.ncores,
+        "transpose_input": False,
+        "transpose_group": False,
+    }
+    script = "file://../scripts/stats/DiffCoexpr.R"
+
+
+class MetaPvalue(Proc):
+    """Calulation of meta p-values.
+
+    If there is only one input file, only the p-value adjustment will be performed.
+
+    Input:
+        infiles: The input files. Each file is a tab-delimited file with multiple
+            columns. There should be ID column(s) to match the rows in other files and
+            p-value column(s) to be combined. The records will be full-joined by ID.
+            When only one file is provided, only the pvalue adjustment will be
+            performed when `envs.padj` is not `none`, otherwise the input file will
+            be copied to `out.outfile`.
+
+    Output:
+        outfile: The output file. It is a tab-delimited file with the first column as
+            the ID and the second column as the combined p-value.
+            ```
+            ID  ID1 ...  Pval   Padj
+            a   x   ...  0.123  0.123
+            b   y   ...  0.123  0.123
+            ...
+            ```
+
+    Envs:
+        id_cols: The column names used in all `in.infiles` as ID columns. Multiple
+            columns can be specified by comma-seperated values. For example, `ID1,ID2`.
+            If `id_expr` is specified, this should be a single column name for the new
+            ID column in each `in.infiles` and the final `out.outfile`.
+        id_exprs: The R expressions for each `in.infiles` to get ID column(s).
+        pval_cols: The column names used in all `in.infiles` as p-value columns.
+            Different columns can be specified by comma-seperated values for each
+            `in.infiles`. For example, `Pval1,Pval2`.
+        method (choice): The method used to calculate the meta-pvalue.
+            - fisher: Fisher's method.
+            - sumlog: Sum of logarithms (same as Fisher's method)
+            - logitp: Logit method.
+            - sumz: Sum of z method (Stouffer's method).
+            - meanz: Mean of z method.
+            - meanp: Mean of p method.
+            - invt: Inverse t method.
+            - sump: Sum of p method (Edgington's method).
+            - votep: Vote counting method.
+            - wilkinsonp: Wilkinson's method.
+            - invchisq: Inverse chi-square method.
+        na: The method to handle NA values. -1 to skip the record. Otherwise NA
+            will be replaced by the given value.
+        padj (choice): The method for p-value adjustment.
+            - none: No p-value adjustment (no Padj column in outfile).
+            - holm: Holm-Bonferroni method.
+            - hochberg: Hochberg method.
+            - hommel: Hommel method.
+            - bonferroni: Bonferroni method.
+            - BH: Benjamini-Hochberg method.
+            - BY: Benjamini-Yekutieli method.
+            - fdr: FDR correction method.
+    """
+    input = "infiles:files"
+    output = "outfile:file:{{in.infiles | first | stem}}.metapval.txt"
+    lang = config.lang.rscript
+    envs = {
+        "id_cols": None,
+        "id_exprs": None,
+        "pval_cols": None,
+        "method": "fisher",
+        "na": -1,
+        "padj": "none",
+    }
+    script = "file://../scripts/stats/MetaPvalue.R"