biopipen 0.21.0__py3-none-any.whl → 0.34.26__py3-none-any.whl

biopipen/ns/protein.py

@@ -0,0 +1,183 @@
+"""Protein-related processes."""
+from ..core.proc import Proc
+from ..core.config import config
+
+
+class Prodigy(Proc):
+    """Prediction of binding affinity of protein-protein complexes based on
+    intermolecular contacts using Prodigy.
+
+    See <https://rascar.science.uu.nl/prodigy/> and
+    <https://github.com/haddocking/prodigy>.
+
+    `prodigy-prot` must be installed under the given python of `proc.lang`.
+
+    Input:
+        infile: The structure file in PDB or mmCIF format.
+
+    Output:
+        outfile: The output file generated by Prodigy.
+        outdir: The output directory containing all output files.
+
+    Envs:
+        distance_cutoff (type=float): The distance cutoff to calculate intermolecular
+            contacts.
+        acc_threshold (type=float): The accessibility threshold for BSA analysis.
+        temperature (type=float): The temperature (C) for Kd prediction.
+        contact_list (flag): Whether to generate contact list.
+        pymol_selection (flag): Whether output a script to highlight the interface
+            residues in PyMOL.
+        selection (list): The selection of the chains to analyze.
+            `['A', 'B']` will analyze chains A and B.
+            `['A,B', 'C']` will analyze chain A and C; and B and C.
+            `['A', 'B', 'C']` will analyze all combinations of A, B, and C.
+        outtype (choice): Set the format of the output file (`out.outfile`).
+            All three files will be generated. This option only determines which
+            is assigned to `out.outfile`.
+            - raw: The raw output file from prodigy.
+            - json: The output file in JSON format.
+            - tsv: The output file in CSV format.
+    """
+    input = "infile:file"
+    output = [
+        "outfile:file:{{in.infile | stem}}_prodigy/"
+        "{{in.infile | stem}}.{{envs.outtype if envs.outtype != 'raw' else 'out'}}",
+        "outdir:dir:{{in.infile | stem}}_prodigy",
+    ]
+    lang = config.lang.python
+    envs = {
+        "distance_cutoff": 5.5,
+        "acc_threshold": 0.05,
+        "temperature": 25.0,
+        "contact_list": True,
+        "pymol_selection": True,
+        "selection": None,
+        "outtype": "json",
+    }
+    script = "file://../scripts/protein/Prodigy.py"
+
+
+class ProdigySummary(Proc):
+    """Summary of the output from `Prodigy`.
+
+    Input:
+        infiles: The output json file generated by `Prodigy`.
+
+    Output:
+        outdir: The directory of summary files generated by `ProdigySummary`.
+
+    Envs:
+        group (type=auto): The group of the samples for boxplots.
+            If `None`, don't do boxplots.
+            It can be a dict of group names and sample names, e.g.
+            `{"group1": ["sample1", "sample2"], "group2": ["sample3"]}`
+            or a file containing the group information, with the first column
+            being the sample names and the second column being the group names.
+            The file should be tab-delimited with no header.
+    """
+    input = "infiles:files"
+    input_data = lambda ch: [[f"{odir}/_prodigy.tsv" for odir in ch.outdir]]
+    output = "outdir:dir:prodigy_summary"
+    lang = config.lang.rscript
+    envs = {"group": None}
+    script = "file://../scripts/protein/ProdigySummary.R"
+    plugin_opts = {"report": "file://../reports/protein/ProdigySummary.svelte"}
+
+
+class MMCIF2PDB(Proc):
+    """Convert mmCIF or PDBx file to PDB file.
+
+    Using [BeEM](https://github.com/kad-ecoli/BeEM)
+
+    Input:
+        infile: The input mmCIF or PDBx file.
+
+    Output:
+        outfile: The output PDB file.
+            The "outfmt" set to 3 to always output a single PDB file.
+
+    Envs:
+        tool (choice): The tool to use for conversion.
+            - maxit: Use MAXIT.
+            - beem: Use BeEM.
+        maxit: The path to the MAXIT executable.
+        beem: The path to the BeEM executable.
+        <more>: Other options for MAXIT/BeEM.
+            For BeEM, "outfmt" will not be used as it is set to 3.
+    """
+    input = "infile:file"
+    output = "outfile:file:{{in.infile | stem}}.pdb"
+    lang = config.lang.python
+    envs = {
+        "tool": "maxit",
+        "maxit": config.exe.maxit,
+        "beem": config.exe.beem,
+    }
+    script = "file://../scripts/protein/MMCIF2PDB.py"
+
+
+class RMSD(Proc):
+    """Calculate the RMSD between two structures.
+
+    See also https://github.com/charnley/rmsd.
+
+    If the input is in mmCIF format, convert it to PDB first.
+
+    Input:
+        infile1: The first structure file.
+        infile2: The second structure file.
+
+    Output:
+        outfile: The output file containing the RMSD value.
+
+    Envs:
+        beem: The path to the BeEM executable.
+        calculate_rmsd: The path to the calculate_rmsd executable.
+        conv_tool (choice): The tool to use for conversion.
+            - maxit: Use MAXIT.
+            - beem: Use BeEM.
+        ca_only (flag): Whether to calculate RMSD using only C-alpha atoms.
+        duel (choice): How to handle the duel atoms. Default is "keep".
+            - keep: Keep both atoms.
+            - keep_first: Keep the first atom.
+            - keep_last: Keep the last atom.
+            - average: Average the coordinates.
+        reorder (flag): Whether to reorder the atoms in the structures.
+        <more>: Other options for calculate_rmsd.
+    """
+    input = "infile1:file, infile2:file"
+    output = "outfile:file:{{in.infile1 | stem}}-{{in.infile2 | stem}}.rmsd.txt"
+    lang = config.lang.python
+    envs = {
+        "maxit": config.exe.maxit,
+        "beem": config.exe.beem,
+        "calculate_rmsd": config.exe.calculate_rmsd,
+        "conv_tool": "maxit",
+        "ca_only": False,
+        "duel": "keep",
+        "reorder": True,
+    }
+    script = "file://../scripts/protein/RMSD.py"
+
+
+class PDB2Fasta(Proc):
+    """Convert PDB file to FASTA file.
+
+    Input:
+        infile: The input PDB file.
+
+    Output:
+        outfile: The output FASTA file.
+
+    Envs:
+        chains (auto): The chains to extract. A list of chain IDs or separated by
+            commas.
+            If None, extract all chains.
+        wrap (type=int): The number of residues per line in the output FASTA
+            file. Set to 0 to disable wrapping.
+    """
+    input = "infile:file"
+    output = "outfile:file:{{in.infile | stem}}.fasta"
+    lang = config.lang.python
+    envs = {"chains": None, "wrap": 80}
+    script = "file://../scripts/protein/PDB2Fasta.py"

biopipen/ns/regulatory.py

@@ -0,0 +1,290 @@
+"""Provides processes for the regulatory related"""
+
+from ..core.proc import Proc
+from ..core.config import config
+
+
+class MotifScan(Proc):
+    """Scan the input sequences for binding sites using motifs.
+
+    Currently only [fimo](https://meme-suite.org/meme/tools/fimo) from MEME suite
+    is supported, based on the research/comparisons done by the following reference.
+
+    Reference:
+        - [Evaluating tools for transcription factor binding site prediction](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6889335/)
+
+    Input:
+        motiffile: File containing motif names.
+            The file contains the motif and regulator names.
+            The motif names should match the names in the motif database.
+            This file must have a header.
+            If multiple columns are present, it should be delimited by tab.
+        seqfile: File containing sequences in FASTA format.
+
+    Output:
+        outdir: Directory containing the results.
+            Especially `fimo_output.txt` extending from `fimo.tsv`, which contains:
+            1. the results with the regulator information if `envs.regulator_col`
+                is provided, otherwise, the `regulator` columns will be filled with
+                the motif names.
+            2. the original sequence from the fasta file (in.seqfile)
+            3. corrected genomic coordinates if the genomic coordinates are included
+                in the sequence names.
+
+            See also the `Output` section of
+            <https://meme-suite.org/meme/doc/fimo.html>.
+            Note that `--no-pgc` is passed to fimo to not parse the genomic coordinates
+            from the sequence names by fimo. When fimo parses the genomic coordinates,
+            `DDX11L1` in `>DDX11L1::chr1:11869-14412` will be lost.
+            The purpose of this is to keep the sequence names as they are in the output.
+            If the sequence names are in the format of `>NAME::chr1:START-END`, we will
+            correct the coordinates in the output.
+            Also note that it requires meme/fimo v5.5.5+ to do this
+            (where the --no-pgc option is available).
+
+    Envs:
+        tool (choice): The tool to use for scanning.
+            Currently only fimo is supported.
+            - fimo: Use fimo from MEME suite.
+        fimo: The path to fimo binary.
+        motif_col: The column name in the motif file containing the motif names.
+        regulator_col: The column name in the motif file containing the regulator names.
+            Both `motif_col` and `regulator_col` should be the direct column names or
+            the index (1-based) of the columns.
+            If no `regulator_col` is provided, no regulator information is written in
+            the output.
+        notfound (choice): What to do if a motif is not found in the database.
+            - error: Report error and stop the process.
+            - ignore: Ignore the motif and continue.
+        motifdb: The path to the motif database. This is required.
+            It should be in the format of MEME motif database.
+            Databases can be downloaded here: <https://meme-suite.org/meme/doc/download.html>.
+            See also introduction to the databases: <https://meme-suite.org/meme/db/motifs>.
+        cutoff (type=float): The cutoff for p-value to write the results.
+            When `envs.q_cutoff` is set, this is applied to the q-value.
+            This is passed to `--thresh` in fimo.
+        q (flag): Calculate q-value.
+            When `False`, `--no-qvalue` is passed to fimo.
+            The q-value calculation is that of Benjamini and Hochberg (BH) (1995).
+        q_cutoff (flag): Apply `envs.cutoff` to q-value.
+        args (ns): Additional arguments to pass to the tool.
+            - <more>: Additional arguments for fimo.
+                See: <https://meme-suite.org/meme/doc/fimo.html>
+    """  # noqa: E501
+    input = "motiffile:file, seqfile:file"
+    output = "outdir:dir:{{in.motiffile | stem}}.fimo"
+    lang = config.lang.python
+    envs = {
+        "tool": "fimo",
+        "fimo": config.exe.fimo,
+        "motif_col": 1,
+        "regulator_col": None,
+        "notfound": "error",
+        "motifdb": config.tf_motifdb,
+        "cutoff": 1e-4,
+        "q": False,
+        "q_cutoff": False,
+        "args": {},
+    }
+    script = "file://../scripts/regulatory/MotifScan.py"
+
+
+class MotifAffinityTest(Proc):
+    """Test the affinity of motifs to the sequences and the affinity change
+    due the mutations.
+
+    See also <https://simon-coetzee.github.io/motifBreakR> and
+    <https://www.bioconductor.org/packages/release/bioc/vignettes/atSNP/inst/doc/atsnp-vignette.html>
+
+    When using atSNP, motifBreakR is also required to plot the variants and motifs.
+
+    Input:
+        motiffile: File containing motif names.
+            The file contains the motif and regulator names.
+            The motif names should match the names in the motif database.
+            This file must have a header.
+            If multiple columns are present, it should be delimited by tab.
+        varfile: File containing the variants.
+            It could be a VCF file or a BED-like file.
+            If it is a VCF file, it does not need to be indexed. Only records with `PASS` in the `FILTER` column are used.
+            If it is a BED-like file, it should contain the following columns, `chrom`, `start`, `end`, `name`, `score`, `strand`, `ref`, `alt`.
+
+    Output:
+        outdir: Directory containing the results.
+            For motifBreakR, `motifbreakr.txt` will be created. Records with effect `strong`/`weak` are written (`neutral` is not).
+            For atSNP, `atsnp.txt` will be created. Records with p-value (`envs.atsnp_args.p`) < `envs.cutoff` are written.
+
+    Envs:
+        ncores (type=int): The number of cores to use.
+        tool (choice): The tool to use for the test.
+            - motifbreakr: Use motifBreakR.
+            - motifBreakR: Use motifBreakR.
+            - atsnp: Use atSNP.
+            - atSNP: Use atSNP.
+        bcftools: The path to bcftools binary.
+            Used to convert the VCF file to the BED file when the input is a VCF file.
+        motif_col: The column name in the motif file containing the motif names.
+            If this is not provided, `envs.regulator_col` and `envs.regmotifs` are required,
+            which are used to infer the motif names from the regulator names.
+        regulator_col: The column name in the motif file containing the regulator names.
+            Both `motif_col` and `regulator_col` should be the direct column names or
+            the index (1-based) of the columns.
+            If no `regulator_col` is provided, no regulator information is written in
+            the output. Otherwise, the regulator information is written in the output in
+            the `Regulator` column.
+        var_col: The column names in the `in.motiffile` containing the variant information.
+            It has to be matching the names in the `in.varfile`. This is helpful when
+            we only need to test the pairs of variants and motifs in the `in.motiffile`.
+        notfound (choice): What to do if a motif is not found in the database,
+            or a regulator is not found in the regulator-motif mapping (envs.regmotifs)
+            file.
+            - error: Report error and stop the process.
+            - ignore: Ignore the motif and continue.
+        motifdb: The path to the motif database. This is required.
+            It should be in the format of MEME motif database.
+            Databases can be downloaded here: <https://meme-suite.org/meme/doc/download.html>.
+            See also introduction to the databases: <https://meme-suite.org/meme/db/motifs>.
+            [universalmotif](https://github.com/bjmt/universalmotif) is required to read the motif database.
+        genome: The genome assembly.
+            Used to fetch the sequences around the variants by package, for example, `BSgenome.Hsapiens.UCSC.hg19` is required if
+            `hg19`. If it is an organism other than human, please specify the full name of the package, for example, `BSgenome.Mmusculus.UCSC.mm10`.
+        cutoff (type=float): The cutoff for p-value to write the results.
+        devpars (ns): The default device parameters for the plot.
+            - width (type=int): The width of the plot.
+            - height (type=int): The height of the plot.
+            - res (type=int): The resolution of the plot.
+        plot_nvars (type=int): Number of variants to plot.
+            Plot top `<plot_nvars>` variants with the largest `abs(alleleDiff)` (motifBreakR) or smallest p-values (atSNP).
+        plots (type=json): Specify the details for the plots.
+            When specified, `plot_nvars` is ignored.
+            The keys are the variant names and the values are the details for the plots, including:
+            devpars: The device parameters for the plot to override the default (envs.devpars).
+            which: An expression passed to `subset(results, subset = ...)` to get the motifs for the variant to plot.
+                Or an integer to get the top `which` motifs.
+                For example, `effect == "strong"` to get the motifs with strong effect in motifBreakR result.
+        regmotifs: The path to the regulator-motif mapping file.
+            It must have header and the columns `Motif` or `Model` for motif names and
+            `TF`, `Regulator` or `Transcription factor`  for regulator names.
+        motifbreakr_args (ns): Additional arguments to pass to motifBreakR.
+            - method (choice): The method to use.
+                See details of <https://rdrr.io/bioc/motifbreakR/man/motifbreakR.html>
+                and <https://simon-coetzee.github.io/motifBreakR/#methods>.
+                - default: Use the default method.
+                - log: Use the standard summation of log probabilities
+                - ic: Use information content
+                - notrans: Use the default method without transformation
+        atsnp_args (ns): Additional arguments to pass to atSNP.
+            - padj_cutoff (flag): The `envs.cutoff` will be applied to the adjusted p-value.
+                Only works for `atSNP`.
+            - padj (choice): The method to adjust the p-values.
+                Only works for `atSNP`
+                - holm: Holm's method
+                - hochberg: Hochberg's method
+                - hommel: Hommel's method
+                - bonferroni: Bonferroni method
+                - BH: Benjamini & Hochberg's method
+                - BY: Benjamini & Yekutieli's method
+                - fdr: False discovery rate
+                - none: No adjustment
+            - p (choice): Which p-value to use for adjustment and cutoff.
+                - pval_ref: p-value for the reference allele affinity score.
+                - pval_snp: p-value for the SNP allele affinity score.
+                - pval_cond_ref: and
+                - pval_cond_snp: conditional p-values for the affinity scores of the reference and SNP alleles.
+                - pval_diff: p-value for the affinity score change between the two alleles.
+                - pval_rank: p-value for the rank test between the two alleles.
+    """  # noqa: E501
+    input = "motiffile:file, varfile:file"
+    output = "outdir:dir:{{in.motiffile | stem}}.{{envs.tool | lower}}"
+    lang = config.lang.rscript
+    envs = {
+        "ncores": config.misc.ncores,
+        "tool": "atsnp",
+        "bcftools": config.exe.bcftools,
+        "motif_col": None,
+        "regulator_col": None,
+        "var_col": None,
+        "notfound": "error",
+        "motifdb": config.ref.tf_motifdb,
+        "regmotifs": config.ref.tf_motifs,
+        "genome": config.ref.genome,
+        "cutoff": 0.05,
+        "devpars": {"width": None, "height": None, "res": 100},
+        "plot_nvars": 10,
+        "plots": {},
+        "motifbreakr_args": {"method": "default"},
+        "atsnp_args": {"padj_cutoff": True, "padj": "BH", "p": "pval_diff"},
+    }
+    script = "file://../scripts/regulatory/MotifAffinityTest.R"
+
+
+class VariantMotifPlot(Proc):
+    """A plot with a genomic region surrounding a genomic variant, and
+    potentially disrupted motifs.
+
+    Currently only SNVs are supported.
+
+    Input:
+        infile: File containing the variants and motifs.
+            It is a TAB-delimited file with the following columns:
+            - chrom: The chromosome of the SNV. Alias: chr, seqnames.
+            - start: The start position of the SNV, no matter 0- or 1-based.
+            - end: The end position of the SNV, which will be used as the position of the SNV.
+            - strand: Indicating the direction of the surrounding sequence matching the motif.
+            - SNP_id: The name of the SNV.
+            - REF: The reference allele of the SNV.
+            - ALT: The alternative allele of the SNV.
+            - providerId: The motif id. It can be specified by `envs.motif_col`.
+            - providerName: The name of the motif provider. Optional.
+            - Regulator: The regulator name. Optional, can be specified by `envs.regulator_col`.
+            - motifPos: The position of the motif, relative to the position of the SNV.
+                For example, '-8, 4' means the motif is 8 bp upstream and 4 bp downstream of the SNV.
+
+    Envs:
+        genome: The genome assembly.
+            Used to fetch the sequences around the variants by package, for example, `BSgenome.Hsapiens.UCSC.hg19` is required if
+            `hg19`. If it is an organism other than human, please specify the full name of the package, for example, `BSgenome.Mmusculus.UCSC.mm10`.
+        motifdb: The path to the motif database. This is required.
+            It should be in the format of MEME motif database.
+            Databases can be downloaded here: <https://meme-suite.org/meme/doc/download.html>.
+            See also introduction to the databases: <https://meme-suite.org/meme/db/motifs>.
+            [universalmotif](https://github.com/bjmt/universalmotif) is required to read the motif database.
+        motif_col: The column name in the motif file containing the motif names.
+            If this is not provided, `envs.regulator_col` and `envs.regmotifs` are required,
+            which are used to infer the motif names from the regulator names.
+        regulator_col: The column name in the motif file containing the regulator names.
+            Both `motif_col` and `regulator_col` should be the direct column names or
+            the index (1-based) of the columns.
+            If no `regulator_col` is provided, no regulator information is written in
+            the output. Otherwise, the regulator information is written in the output in
+            the `Regulator` column.
+        regmotifs: The path to the regulator-motif mapping file.
+            It must have header and the columns `Motif` or `Model` for motif names and
+            `TF`, `Regulator` or `Transcription factor`  for regulator names.
+        notfound (choice): What to do if a motif is not found in the database,
+            or a regulator is not found in the regulator-motif mapping (envs.regmotifs)
+            file.
+            - error: Report error and stop the process.
+            - ignore: Ignore the motif and continue.
+        devpars (ns): The default device parameters for the plot.
+            - width (type=int): The width of the plot.
+            - height (type=int): The height of the plot.
+            - res (type=int): The resolution of the plot.
+        plot_vars (type=auto): The variants (SNP_id) to plot.
+            A list of variant names to plot or a string with the variant names separated by comma.
+            When not specified, all variants are plotted.
+    """  # noqa: E501
+    input = "infile:file"
+    output = "outdir:dir:{{in.infile | stem}}.vmplots"
+    lang = config.lang.rscript
+    envs = {
+        "genome": config.ref.genome,
+        "motifdb": config.ref.tf_motifdb,
+        "motif_col": "providerId",
+        "regulator_col": None,
+        "regmotifs": config.ref.tf_motifs,
+        "notfound": "error",
+        "devpars": {"width": 800, "height": None, "res": 100},
+        "plot_vars": None,
+    }
+    script = "file://../scripts/regulatory/VariantMotifPlot.R"

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"