biopipen 0.28.1__py3-none-any.whl → 0.29.0__py3-none-any.whl

biopipen/ns/snp.py

@@ -5,9 +5,9 @@ from ..core.config import config
 
 
 class PlinkSimulation(Proc):
-    """Simulate SNPs using PLINK v1.9
+    """Simulate SNPs using PLINK v2
 
-    See also <https://www.cog-genomics.org/plink/1.9/input#simulate> and
+    See also <https://www.cog-genomics.org/plink/2.0/input#simulate> and
     <https://pwwang.github.io/biopipen/api/biopipen.ns.snp/#biopipen.ns.snp.PlinkSimulation>
 
     Input:
@@ -27,7 +27,7 @@ class PlinkSimulation(Proc):
         nsnps (type=int): Number of SNPs to simulate
         ncases (type=int): Number of cases to simulate
         nctrls (type=int): Number of controls to simulate
-        plink: Path to PLINK v1.9
+        plink: Path to PLINK v2
         seed (type=int): Random seed. If not set, seed will not be set.
         label: Prefix label for the SNPs.
         prevalence  (type=float): Disease prevalence.
@@ -37,7 +37,7 @@ class PlinkSimulation(Proc):
         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>.
+            - <more>: see <https://www.cog-genomics.org/plink/2.0/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
@@ -93,11 +93,15 @@ class MatrixEQTL(Proc):
 
     Envs:
         model (choice): The model to use.
-            - `linear`: Linear model
-            - `modelLINEAR`: Same as `linear`
-            - `anova`: ANOVA model
-            - `modelANOVA`: Same as `anova`
+            - linear: Linear model
+            - modelLINEAR: Same as `linear`
+            - anova: ANOVA model
+            - modelANOVA: Same as `anova`
         pval (type=float): P-value threshold for eQTLs
+        match_samples (flag): Match samples in the genotype and expression matrices.
+            If True, an error will be raised if samples from `in.geno`, `in.expr`,
+            and `in.cov` (if provided) are not the same.
+            If False, common samples will be used to subset the matrices.
         transp (type=float): P-value threshold for trans-eQTLs.
             If cis-eQTLs are not enabled (`snppos` and `genepos` are not set),
             this defaults to 1e-5.
@@ -126,6 +130,7 @@ class MatrixEQTL(Proc):
     envs = {
         "model": "linear",
         "pval": 1e-3,
+        "match_samples": False,
         "transp": None,
         "fdr": False,
         "snppos": None,
@@ -136,3 +141,506 @@ class MatrixEQTL(Proc):
         "transpose_cov": False,
     }
     script = "file://../scripts/snp/MatrixEQTL.R"
+
+
+class PlinkFromVcf(Proc):
+    """Convert VCF to PLINK format.
+
+    The PLINK format consists of 3 files: `.bed`, `.bim`, and `.fam`.
+
+    Requires PLINK v2
+
+    TODO:
+        Handle sex when sex chromosomes are included.
+
+    Input:
+        invcf: VCF file
+
+    Output:
+        outdir: Output directory containing the PLINK files
+
+    Envs:
+        plink: Path to PLINK v2
+        tabix: Path to tabix
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        vcf_half_call (choice): The current VCF standard does not specify
+            how '0/.' and similar GT values should be interpreted.
+            - error: error out and reports the line number of the anomaly
+            - e: alias for `error`
+            - haploid: treat half-calls as haploid/homozygous
+            - h: alias for `haploid`
+            - missing: treat half-calls as missing
+            - m: alias for `missing`
+            - reference: treat the missing part as reference
+            - r: alias for `reference`
+        double_id (flag): set both FIDs and IIDs to the VCF/BCF sample ID.
+        vcf_filter (auto): skip variants which failed one or more filters tracked
+            by the FILTER field.
+            If True, only FILTER with `PASS` or `.` will be kept.
+            Multiple filters can be specified by separating them with space or
+            as a list.
+        vcf_idspace_to: convert all spaces in sample IDs to this character.
+        set_missing_var_ids: update variant IDs using a template string,
+            with a '@' where the chromosome code should go, and a '#' where the
+            base-pair position belongs. You can also specify `\$r` and `\$a` for
+            the reference and alternate alleles, respectively.
+            See <https://www.cog-genomics.org/plink/2.0/data#set_all_var_ids>
+        max_alleles (type=int): Maximum number of alleles per variant.
+        <more>: see <https://www.cog-genomics.org/plink/2.0/> for more options.
+            Note that `_` will be replaced by `-` in the argument names.
+    """  # noqa: E501
+    input = "invcf:file"
+    output = "outdir:dir:{{in.invcf | regex_replace: '\\.gz$', '' | stem}}"
+    lang = config.lang.python
+    envs = {
+        "plink": config.exe.plink2,
+        "tabix": config.exe.tabix,
+        "ncores": config.misc.ncores,
+        "vcf_half_call": "missing",
+        "double_id": True,
+        "vcf_filter": True,
+        "vcf_idspace_to": "_",
+        "set_missing_var_ids": "@_#",
+        "max_alleles": 2,
+    }
+    script = "file://../scripts/snp/PlinkFromVcf.py"
+
+
+class Plink2GTMat(Proc):
+    """Convert PLINK files to genotype matrix.
+
+    Requires PLINK v2. The .raw/.traw file is generated by plink and then transformed
+    to a genotype matrix file.
+    See <https://www.cog-genomics.org/plink/2.0/formats#raw> and
+    <https://www.cog-genomics.org/plink/2.0/formats#traw> for more information.
+
+    The allelic dosage is used as the values of genotype matrix.
+    "--keep-allele-order" is used to keep the allele order consistent with the
+    reference allele first.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outfile: Genotype matrix file with rows representing SNPs and columns
+            representing samples if `envs.transpose` is `False`.
+
+    Envs:
+        plink: Path to PLINK v2.0
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        transpose (flag): If set, the genotype matrix (`out.outfile`) is transposed.
+        samid: what to use as sample ID.
+            Placeholders include `{fid}` and `{iid}` for family and individual IDs,
+            respectively.
+        varid: what to use as variant ID.
+            Placeholders include `{chr}`, `{pos}`, `{rs}`, `{ref}`, and `{alt}` for
+            chromosome, position, rsID, reference allele, and alternate allele,
+            respectively.
+        trans_chr: A dictionary to translate chromosome numbers to chromosome names.
+        missing_id: what to use as the rs if missing.
+    """
+    input = "indir:dir"
+    output = "outfile:file:{{in.indir | stem}}-gtmat.txt"
+    lang = config.lang.python
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "transpose": False,
+        "samid": "{fid}_{iid}",
+        "varid": "{chr}_{pos}_{varid}_{ref}_{alt}",
+        "trans_chr": {"23": "X", "24": "Y", "25": "XY", "26": "M"},
+        "missing_id": "NA",
+    }
+    script = "file://../scripts/snp/Plink2GTMat.py"
+
+
+class PlinkIBD(Proc):
+    """Run PLINK IBD analysis (identity by descent)
+
+    See also <https://www.cog-genomics.org/plink/1.9/ibd>
+    This has to run with PLINK v1.9. Plink v2 does not support IBD analysis yet.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outdir: Output file containing the IBD results.
+            Including [`.genome`](https://www.cog-genomics.org/plink/2.0/formats#genome)
+            file for the original IBD report from PLINK, and `.ibd.png` for the
+            heatmap of `PI_HAT` values.
+
+    Envs:
+        plink: Path to PLINK v1.9
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        highld: High LD regions to be excluded from the analysis.
+            If not set, no regions will be excluded.
+        samid: what to use as sample ID.
+            Placeholders include `{fid}` and `{iid}` for family and individual IDs,
+            respectively
+        indep (type=auto): LD pruning parameters. Either a list of numerics or a string
+            concatenated by `,` to specify
+            1) consider a window of N SNPs (e.g. 50),
+            2) calculate LD between each pair of SNPs in the window (e.g. 5),
+            3) remove one of a pair of SNPs if the LD is greater than X (e.g. 0.2).
+        pihat (type=float): PI_HAT threshold for IBD analysis.
+            See also <https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5007749/>
+        plot (flag): If set, plot the heatmap of `PI_HAT` values.
+        anno: The annotation file for the samples, used to plot on the heatmap.
+            Names must match the ones that are transformed by `args.samid`.
+        seed (type=int): Random seed for the analysis.
+        devpars (ns): The device parameters for the plot.
+            - width (type=int): Width of the plot
+            - height (type=int): Height of the plot
+            - res (type=int): Resolution of the plot
+    """
+    input = "indir:dir"
+    output = "outdir:dir:{{in.indir | stem}}.ibd"
+    lang = config.lang.rscript
+    envs = {
+        "plink": config.exe.plink,
+        "ncores": config.misc.ncores,
+        "highld": None,
+        "samid": "{fid}_{iid}",
+        "indep": [50, 5, 0.2],
+        "pihat": 0.1875,
+        "plot": True,
+        "anno": None,
+        "seed": 8525,
+        "devpars": {"width": 1000, "height": 1000, "res": 100},
+    }
+    script = "file://../scripts/snp/PlinkIBD.R"
+    plugin_opts = {"report": "file://../reports/snp/PlinkIBD.svelte"}
+
+
+class PlinkHWE(Proc):
+    """Hardy-Weinberg Equilibrium report and filtering
+
+    See also <https://www.cog-genomics.org/plink/2.0/basic_stats#hardy>
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outdir: Output file containing the HWE results.
+            Including [`.hwe`](https://www.cog-genomics.org/plink/2.0/formats#hwe)
+            file for the original HWE report from PLINK and
+            `.hardy.fail` for the variants that failed the HWE test.
+            It also includes binary files `.bed`, `.bim`, and `.fam`
+
+    Envs:
+        plink: Path to PLINK v2
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        cutoff (type=float): P-value cutoff for HWE test
+        plot (flag): If set, plot the distribution of HWE p-values.
+        devpars (ns): The device parameters for the plot.
+            - width (type=int): Width of the plot
+            - height (type=int): Height of the plot
+            - res (type=int): Resolution of the plot
+    """
+    input = "indir:dir"
+    output = "outdir:dir:{{in.indir | stem}}.hwe"
+    lang = config.lang.rscript
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "cutoff": 1e-5,
+        "plot": True,
+        "devpars": {"width": 1000, "height": 800, "res": 100},
+    }
+    script = "file://../scripts/snp/PlinkHWE.R"
+    plugin_opts = {"report": "file://../reports/snp/PlinkHWE.svelte"}
+
+
+class PlinkHet(Proc):
+    """Calculation of sample heterozygosity.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outdir: Output file containing the heterozygosity results.
+            Including [`.het`](https://www.cog-genomics.org/plink/2.0/formats#het)
+            file for the original heterozygosity report from PLINK and
+            `.het.fail` for the samples that failed the heterozygosity test.
+            It also includes binary files `.bed`, `.bim`, and `.fam`
+
+    Envs:
+        plink: Path to PLINK v2, at least v2.00a5.10
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        cutoff (type=float): Heterozygosity cutoff, samples with heterozygosity
+            beyond `mean - cutoff * sd` or `mean + cutoff * sd` will be considered
+            as outliers.
+        plot (flag): If set, plot the distribution of heterozygosity values.
+        devpars (ns): The device parameters for the plot.
+            - width (type=int): Width of the plot
+            - height (type=int): Height of the plot
+            - res (type=int): Resolution of the plot
+    """
+    input = "indir:dir"
+    output = "outdir:dir:{{in.indir | stem}}.het"
+    lang = config.lang.rscript
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "cutoff": 3.0,
+        "plot": True,
+        "devpars": {"width": 1000, "height": 800, "res": 100},
+    }
+    script = "file://../scripts/snp/PlinkHet.R"
+    plugin_opts = {"report": "file://../reports/snp/PlinkHet.svelte"}
+
+
+class PlinkCallRate(Proc):
+    """Calculation of call rate for the samples and variants.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outdir: Output file containing the call rate results.
+            Including [`.imiss`](https://www.cog-genomics.org/plink/2.0/formats#imiss)
+            file for missing calls for samples,
+            [`.lmiss`](https://www.cog-genomics.org/plink/2.0/formats#lmiss) for
+            missing calls for variants, `.samplecr.fail` for the samples fail
+            sample call rate cutoff (`args.samplecr`), and `.varcr.fail` for the SNPs
+            fail snp call rate cutoff (`args.varcr`).
+            It also includes binary files `.bed`, `.bim`, and `.fam`.
+
+    Envs:
+        plink: Path to PLINK v2
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        samplecr (type=float): Sample call rate cutoff
+        varcr (type=float): Variant call rate cutoff
+        max_iter (type=int): Maximum number of iterations to run the call rate
+            calculation.
+            Since the sample and variant call rates are affected by each other,
+            it may be necessary to iterate the calculation to get the stable results.
+        plot (flag): If set, plot the distribution of call rates.
+        devpars (ns): The device parameters for the plot.
+            - width (type=int): Width of the plot
+            - height (type=int): Height of the plot
+            - res (type=int): Resolution of the plot
+    """
+    input = "indir:dir"
+    output = "outdir:dir:{{in.indir | stem}}.callrate"
+    lang = config.lang.rscript
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "samplecr": 0.95,
+        "varcr": 0.95,
+        "max_iter": 3,
+        "plot": True,
+        "devpars": {"width": 1000, "height": 800, "res": 100},
+    }
+    script = "file://../scripts/snp/PlinkCallRate.R"
+    plugin_opts = {"report": "file://../reports/snp/PlinkCallRate.svelte"}
+
+
+class PlinkFilter(Proc):
+    """Filter samples and variants for PLINK files.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+        samples_file: File containing the sample IDs.
+        variants_file: File containing the variant IDs or regions.
+
+    Output:
+        outdir: Output directory containing the filtered PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Envs:
+        plink: Path to PLINK v2
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        samples (auto): Sample IDs.
+            If both FID and IID should be provided and separatedby `/`. Otherwise,
+            assuming the same FID and IID.
+            A list of sample IDs or string concatenated by `,`.
+            If either `in.samples_file` or `envs.samples_file` is set,
+            this will be ignored.
+        variants (auto): Variant IDs.
+            A list of variant IDs or string concatenated by `,`.
+            If either `in.variants_file` or `envs.variants_file` is set,
+            this will be ignored.
+        samples_file: File containing the sample IDs.
+            If `in.samples_file` is set, this will be ignored.
+        variants_file: File containing the variant IDs.
+            If `in.variants_file` is set, this will be ignored.
+        keep (flag): Use `samples`/`variants`/`samples_file`/`variants_file` to
+            only keep the specified samples/variants, instead of removing them.
+        vfile_type (choice): The type of the variants file.
+            - id: Variant IDs
+            - bed0: 0-based BED file
+            - bed1: 1-based BED file
+        chr: Chromosome to keep.
+            For example, `1-4 22 XY` will keep chromosomes 1 to 4, 22, and XY.
+        not_chr: Chromosome to remove.
+            For example, `1-4 22 XY` will remove chromosomes 1 to 4, 22, and XY.
+        autosome (flag): Excludes all unplaced and non-autosomal variants
+        autosome_xy (flag): Does `autosome` but does not exclude the pseudo-autosomal
+            region of X.
+        snps_only (auto): Excludes all variants with one or more multi-character
+            allele codes. With 'just-acgt', variants with single-character allele codes
+            outside of {'A', 'C', 'G', 'T', 'a', 'c', 'g', 't', <missing code>}
+            are also excluded.
+    """
+    input = [
+        "indir:dir",
+        "samples_file:file",
+        "variants_file:file",
+    ]
+    output = "outdir:dir:{{in.indir | stem}}.filtered"
+    lang = config.lang.python
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "samples": None,
+        "variants": None,
+        "samples_file": None,
+        "variants_file": None,
+        "keep": False,
+        "vfile_type": "id",
+        "chr": None,
+        "not_chr": None,
+        "autosome": False,
+        "autosome_xy": False,
+        "snps_only": False,
+    }
+    script = "file://../scripts/snp/PlinkFilter.py"
+
+
+class PlinkFreq(Proc):
+    """Calculate allele frequencies for the variants.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Output:
+        outdir: Output file containing the allele frequency results.
+            By default, it includes
+            [`.afreq`](https://www.cog-genomics.org/plink/2.0/formats#afreq)
+            file for the allele frequency report from PLINK.
+            Modifiers can be added to change this behavior.
+            See `envs.modifier` for more information.
+            When `envs.filter != no`, it also includes binary files `.bed`, `.bim`,
+            and `.fam` after filtering with `envs.cutoff`.
+
+    Envs:
+        plink: Path to PLINK v2
+        ncores (type=int): Number of cores/threads to use, will pass to plink
+            `--threads` option
+        modifier (choice): The modifier of `--freq` to control the output behavior.
+            - none: No modifier, only the `.afreq` file will be generated.
+                `MAF` (minor allele frequency) will be added in addition to the
+                `REF_FREQ` and `ALT1_FREQ` columns. Check `.afreqx` for the added
+                columns.
+            - counts: write allele count report to `.acount`.
+                See <https://www.cog-genomics.org/plink/2.0/formats#afreq>.
+                `ALT1`, `ALT1_CT`, and `REF_CT` are added. Check `.acountx` for
+                the added columns.
+            - x: write genotype count report to `.gcount`
+                Like `--freqx` in v1.9, `--geno-counts` will be run to generate
+                the genotype counts.
+                `ALT1`, `HET_REF_ALT1_CT`, and `HOM_ALT1_CT` are added. Check
+                `.gcountx` for the added columns.
+        gz (flag): If set, compress the output files.
+        cutoff (auto): Cutoffs to mark or filter the variants.
+            If a float is given, default column will be used based on the modifier.
+            For `modifier="none"`, it defaults to `MAF`.
+            For `modifier="counts"`, it defaults to `ALT1_CT`.
+            For `modifier="x"`, it defaults to `HOM_ALT1_CT`.
+            Or this could be a dictionary to specify the column names and cutoffs.
+            For example, `{"MAF": 0.05}`.
+        filter (auto): The direction of filtering variants based on `cutoff`.
+            If a single value is given, it will apply to all columns provided in
+            `cutoff`. If a dictionary is given, it will apply to the corresponding
+            column. If a column cannot be found in the dictionary, it defaults to
+            `no`.
+            no: Do not filter variants (no binary files are generated in outdir).
+            gt: Filter variants with MAF greater than `cutoff`.
+            lt: Filter variants with MAF less than `cutoff`.
+            ge: Filter variants with MAF greater than or equal to `cutoff`.
+            le: Filter variants with MAF less than or equal to `cutoff`.
+        plot (flag): If set, plot the distribution of allele frequencies.
+        devpars (ns): The device parameters for the plot.
+            - width (type=int): Width of the plot
+            - height (type=int): Height of the plot
+            - res (type=int): Resolution of the plot
+    """
+    input = "indir:dir"
+    output = "outdir:dir:{{in.indir | stem}}.freq"
+    lang = config.lang.rscript
+    envs = {
+        "plink": config.exe.plink2,
+        "ncores": config.misc.ncores,
+        "modifier": "none",
+        "gz": False,
+        "cutoff": {},
+        "filter": {},
+        "plot": True,
+        "devpars": {"width": 1000, "height": 800, "res": 100},
+    }
+    script = "file://../scripts/snp/PlinkFreq.R"
+    plugin_opts = {"report": "file://../reports/snp/PlinkFreq.svelte"}
+
+
+class PlinkUpdateName(Proc):
+    """Update variant names in PLINK files.
+
+    See also <https://www.cog-genomics.org/plink/2.0/data#update_map>.
+
+    Input:
+        indir: Input directory containing the PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+        namefile: File containing the variant names to update.
+            Either a file containing two columns, the first column is the old
+            variant name, and the second column is the new variant name.
+            Or a VCF file containing the variant names to update.
+            When a VCF file is given, the chromosome, position, and reference and
+            alternate alleles will be used to match the variants.
+
+    Output:
+        outdir: Output directory containing the updated PLINK files.
+            Including `.bed`, `.bim`, and `.fam` files
+
+    Envs:
+        ncores: Number of cores/threads to use, will pass to plink `--threads` option
+        plink: Path to PLINK v2
+        bcftools: Path to bcftools
+        match_alt (choice): How to match alternate alleles when `in.namefile`
+            is a VCF file.
+            - exact: Matches alternate alleles exactly.
+            - all: Matches alternate alleles regardless of the order.
+                `chr1:100:A:T,G` matches `chr1:100:A:G,T` or `chr1:100:A:T,G`.
+            - any: Matches any alternate allele.
+                For example, `chr1:100:A:T,G` matches `chr1:100:A:G,C`
+            - first_included: Matches when the first allele is included.
+                For example, `chr1:100:A:T,G` matches `chr1:100:A:C,T`.
+            - first: Match first alternate allele
+                For example, `chr1:100:A:T,G` matches `chr1:100:A:T`.
+            - none: Do not match alternate alleles
+    """
+    input = "indir:dir, namefile:file"
+    output = "outdir:dir:{{in.indir | stem}}.newnames"
+    lang = config.lang.python
+    envs = {
+        "ncores": config.misc.ncores,
+        "plink": config.exe.plink2,
+        "bcftools": config.exe.bcftools,
+        "match_alt": "exact",
+    }
+    script = "file://../scripts/snp/PlinkUpdateName.py"

biopipen/ns/stats.py

@@ -275,8 +275,10 @@ class MetaPvalue(Proc):
 
     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
+            columns can be specified by comma-seperated values. For example, `ID1,ID2`,
+            where `ID1` is the ID column in the first file and `ID2` is the ID column
+            in the second file.
+            If `id_exprs` 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.
@@ -294,6 +296,8 @@ class MetaPvalue(Proc):
             - votep: Vote counting method.
             - wilkinsonp: Wilkinson's method.
             - invchisq: Inverse chi-square method.
+        keep_single (flag): Whether to keep the original p-value when there is only one
+            p-value.
         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.
@@ -315,6 +319,74 @@ class MetaPvalue(Proc):
         "pval_cols": None,
         "method": "fisher",
         "na": -1,
+        "keep_single": True,
         "padj": "none",
     }
     script = "file://../scripts/stats/MetaPvalue.R"
+
+
+class MetaPvalue1(Proc):
+    """Calulation of meta p-values.
+
+    Unlike `MetaPvalue`, this process only accepts one input file.
+
+    The p-values will be grouped by the ID columns and combined by the selected method.
+
+    Input:
+        infile: The input file.
+            The file is a tab-delimited file with multiple
+            columns. There should be ID column(s) to group the rows where
+            p-value column(s) to be combined.
+
+    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 `in.infile` as ID columns. Multiple
+            columns can be specified by comma-seperated values. For example, `ID1,ID2`.
+        pval_col: The column name used in `in.infile` as p-value column.
+        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.
+        keep_single (flag): Whether to keep the original p-value when there is only one
+            p-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 = "infile:file"
+    output = "outfile:file:{{in.infile | stem}}.metapval.txt"
+    lang = config.lang.rscript
+    envs = {
+        "id_cols": None,
+        "pval_col": None,
+        "method": "fisher",
+        "na": -1,
+        "keep_single": True,
+        "padj": "none",
+    }
+    script = "file://../scripts/stats/MetaPvalue1.R"