consenrich 0.7.4b2__cp312-cp312-macosx_10_13_x86_64.whl

consenrich/consenrich.py

@@ -0,0 +1,1381 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import argparse
+import glob
+import logging
+import pprint
+import os
+from pathlib import Path
+from collections.abc import Mapping
+from typing import List, Optional, Tuple, Dict, Any, Union
+import shutil
+import subprocess
+import sys
+import numpy as np
+import pandas as pd
+import pysam
+import pywt
+import yaml
+
+import consenrich.core as core
+import consenrich.misc_util as misc_util
+import consenrich.constants as constants
+import consenrich.detrorm as detrorm
+import consenrich.matching as matching
+
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s - %(module)s.%(funcName)s -  %(levelname)s - %(message)s",
+)
+
+logger = logging.getLogger(__name__)
+
+
+def _loadConfig(
+    configSource: Union[str, Path, Mapping[str, Any]],
+) -> Dict[str, Any]:
+    r"""Load a YAML config from a path or accept an already-parsed mapping.
+
+    If given a dict-like object, just return it.If given a path, try to load as YAML --> dict
+    If given a path, try to load as YAML --> dict
+
+    """
+    if isinstance(configSource, Mapping):
+        configData = configSource
+    elif isinstance(configSource, (str, Path)):
+        with open(configSource, "r") as fileHandle:
+            configData = yaml.safe_load(fileHandle) or {}
+    else:
+        raise TypeError("`config` must be a path or a mapping/dict.")
+
+    if not isinstance(configData, Mapping):
+        raise TypeError("Top-level YAML must be a mapping/object.")
+    return configData
+
+
+def _cfgGet(
+    configMap: Mapping[str, Any],
+    dottedKey: str,
+    defaultVal: Any = None,
+) -> Any:
+    r"""Support both dotted keys and yaml/dict-style nested access for configs."""
+
+    # e.g., inputParams.bamFiles
+    if dottedKey in configMap:
+        return configMap[dottedKey]
+
+    # e.g.,
+    # inputParams:
+    #   bamFiles: [...]
+    currentVal: Any = configMap
+    for keyPart in dottedKey.split("."):
+        if isinstance(currentVal, Mapping) and keyPart in currentVal:
+            currentVal = currentVal[keyPart]
+        else:
+            return defaultVal
+    return currentVal
+
+
+def _listOrEmpty(list_):
+    if list_ is None:
+        return []
+    return list_
+
+
+def _getMinR(configMap, numBams: int) -> float:
+    fallbackMinR: float = 1.0
+    try:
+        rawVal = _cfgGet(configMap, "observationParams.minR", None)
+        return float(rawVal) if rawVal is not None else fallbackMinR
+    except (TypeError, ValueError, KeyError):
+        logger.warning(
+            f"Invalid or missing 'observationParams.minR' in config. Using `{fallbackMinR}`."
+        )
+        return fallbackMinR
+
+
+def checkControlsPresent(inputArgs: core.inputParams) -> bool:
+    """Check if control BAM files are present in the input arguments.
+
+    :param inputArgs: core.inputParams object
+    :return: True if control BAM files are present, False otherwise.
+    """
+    return (
+        bool(inputArgs.bamFilesControl)
+        and isinstance(inputArgs.bamFilesControl, list)
+        and len(inputArgs.bamFilesControl) > 0
+    )
+
+
+def getReadLengths(
+    inputArgs: core.inputParams,
+    countingArgs: core.countingParams,
+    samArgs: core.samParams,
+) -> List[int]:
+    r"""Get read lengths for each BAM file in the input arguments.
+
+    :param inputArgs: core.inputParams object containing BAM file paths.
+    :param countingArgs: core.countingParams object containing number of reads.
+    :param samArgs: core.samParams object containing SAM thread and flag exclude parameters.
+    :return: List of read lengths for each BAM file.
+    """
+    if not inputArgs.bamFiles:
+        raise ValueError(
+            "No BAM files provided in the input arguments."
+        )
+
+    if (
+        not isinstance(inputArgs.bamFiles, list)
+        or len(inputArgs.bamFiles) == 0
+    ):
+        raise ValueError("bam files list is empty")
+
+    return [
+        core.getReadLength(
+            bamFile,
+            countingArgs.numReads,
+            1000,
+            samArgs.samThreads,
+            samArgs.samFlagExclude,
+        )
+        for bamFile in inputArgs.bamFiles
+    ]
+
+
+def checkMatchingEnabled(matchingArgs: core.matchingParams) -> bool:
+    matchingEnabled = (
+        (matchingArgs.templateNames is not None)
+        and isinstance(matchingArgs.templateNames, list)
+        and len(matchingArgs.templateNames) > 0
+    )
+    matchingEnabled = (
+        matchingEnabled
+        and (matchingArgs.cascadeLevels is not None)
+        and isinstance(matchingArgs.cascadeLevels, list)
+        and len(matchingArgs.cascadeLevels) > 0
+    )
+    return matchingEnabled
+
+
+def getEffectiveGenomeSizes(
+    genomeArgs: core.genomeParams, readLengths: List[int]
+) -> List[int]:
+    r"""Get effective genome sizes for the given genome name and read lengths.
+    :param genomeArgs: core.genomeParams object
+    :param readLengths: List of read lengths for which to get effective genome sizes.
+    :return: List of effective genome sizes corresponding to the read lengths.
+    """
+    genomeName = genomeArgs.genomeName
+    if not genomeName or not isinstance(genomeName, str):
+        raise ValueError("Genome name must be a non-empty string.")
+
+    if not isinstance(readLengths, list) or len(readLengths) == 0:
+        raise ValueError(
+            "Read lengths must be a non-empty list. Try calling `getReadLengths` first."
+        )
+    return [
+        constants.getEffectiveGenomeSize(genomeName, readLength)
+        for readLength in readLengths
+    ]
+
+
+def getInputArgs(config_path: str) -> core.inputParams:
+    configData = _loadConfig(config_path)
+
+    def expandWildCards(bamList: List[str]) -> List[str]:
+        expandedList: List[str] = []
+        for bamEntry in bamList:
+            if "*" in bamEntry or "?" in bamEntry or "[" in bamEntry:
+                matchedList = glob.glob(bamEntry)
+                expandedList.extend(matchedList)
+            else:
+                expandedList.append(bamEntry)
+        return expandedList
+
+    bamFilesRaw = (
+        _cfgGet(configData, "inputParams.bamFiles", []) or []
+    )
+    bamFilesControlRaw = (
+        _cfgGet(configData, "inputParams.bamFilesControl", []) or []
+    )
+
+    bamFiles = expandWildCards(bamFilesRaw)
+    bamFilesControl = expandWildCards(bamFilesControlRaw)
+
+    if len(bamFiles) == 0:
+        raise ValueError(
+            "No BAM files provided in the configuration."
+        )
+
+    if (
+        len(bamFilesControl) > 0
+        and len(bamFilesControl) != len(bamFiles)
+        and len(bamFilesControl) != 1
+    ):
+        raise ValueError(
+            "Number of control BAM files must be 0, 1, or the same as number of treatment files"
+        )
+
+    if len(bamFilesControl) == 1:
+        logger.info(
+            f"Only one control given: Using {bamFilesControl[0]} for all treatment files."
+        )
+        bamFilesControl = bamFilesControl * len(bamFiles)
+
+    if not bamFiles or not isinstance(bamFiles, list):
+        raise ValueError("No BAM files found")
+
+    for bamFile in bamFiles:
+        misc_util.checkBamFile(bamFile)
+
+    if bamFilesControl:
+        for bamFile in bamFilesControl:
+            misc_util.checkBamFile(bamFile)
+
+    pairedEndList = misc_util.bamsArePairedEnd(bamFiles)
+    pairedEndConfig: Optional[bool] = _cfgGet(
+        configData, "inputParams.pairedEnd", None
+    )
+    if pairedEndConfig is None:
+        pairedEndConfig = all(pairedEndList)
+        if pairedEndConfig:
+            logger.info("Paired-end BAM files detected")
+        else:
+            logger.info("One or more single-end BAM files detected")
+
+    return core.inputParams(
+        bamFiles=bamFiles,
+        bamFilesControl=bamFilesControl,
+        pairedEnd=pairedEndConfig,
+    )
+
+def getOutputArgs(config_path: str) -> core.outputParams:
+
+    configData = _loadConfig(config_path)
+
+    convertToBigWig_ = _cfgGet(
+        configData, "outputParams.convertToBigWig", True if shutil.which("bedGraphToBigWig") else False
+    )
+
+    roundDigits_ = _cfgGet(
+        configData, "outputParams.roundDigits", 3
+    )
+
+    writeResiduals_ = _cfgGet(
+        configData, "outputParams.writeResiduals", True
+    )
+
+    writeMuncTrace: bool = _cfgGet(
+        configData, "outputParams.writeMuncTrace", False
+    )
+
+    writeStateStd: bool = _cfgGet(
+        configData, "outputParams.writeStateStd", False
+    )
+
+    return core.outputParams(
+        convertToBigWig=convertToBigWig_,
+        roundDigits=roundDigits_,
+        writeResiduals=writeResiduals_,
+        writeMuncTrace=writeMuncTrace,
+        writeStateStd=writeStateStd,
+    )
+
+
+def getGenomeArgs(config_path: str) -> core.genomeParams:
+    configData = _loadConfig(config_path)
+
+    genomeName = _cfgGet(configData, "genomeParams.name", None)
+    genomeLabel = constants.resolveGenomeName(genomeName)
+
+    chromSizesFile: Optional[str] = None
+    blacklistFile: Optional[str] = None
+    sparseBedFile: Optional[str] = None
+    chromosomesList: Optional[List[str]] = None
+
+    excludeChromsList: List[str] = (
+        _cfgGet(configData, "genomeParams.excludeChroms", []) or []
+    )
+    excludeForNormList: List[str] = (
+        _cfgGet(configData, "genomeParams.excludeForNorm", []) or []
+    )
+
+    if genomeLabel:
+        chromSizesFile = constants.getGenomeResourceFile(
+            genomeLabel, "sizes"
+        )
+        blacklistFile = constants.getGenomeResourceFile(
+            genomeLabel, "blacklist"
+        )
+        sparseBedFile = constants.getGenomeResourceFile(
+            genomeLabel, "sparse"
+        )
+
+    chromSizesOverride = _cfgGet(
+        configData, "genomeParams.chromSizesFile", None
+    )
+    if chromSizesOverride:
+        chromSizesFile = chromSizesOverride
+
+    blacklistOverride = _cfgGet(
+        configData, "genomeParams.blacklistFile", None
+    )
+    if blacklistOverride:
+        blacklistFile = blacklistOverride
+
+    sparseOverride = _cfgGet(
+        configData, "genomeParams.sparseBedFile", None
+    )
+    if sparseOverride:
+        sparseBedFile = sparseOverride
+
+    if not chromSizesFile or not os.path.exists(chromSizesFile):
+        raise FileNotFoundError(
+            f"Chromosome sizes file {chromSizesFile} does not exist."
+        )
+
+    chromosomesConfig = _cfgGet(
+        configData, "genomeParams.chromosomes", None
+    )
+    if chromosomesConfig is not None:
+        chromosomesList = chromosomesConfig
+    else:
+        if chromSizesFile:
+            chromosomesFrame = pd.read_csv(
+                chromSizesFile,
+                sep="\t",
+                header=None,
+                names=["chrom", "size"],
+            )
+            chromosomesList = list(chromosomesFrame["chrom"])
+        else:
+            raise ValueError(
+                "No chromosomes provided in the configuration and no chromosome sizes file specified."
+            )
+
+    chromosomesList = [
+        chromName.strip()
+        for chromName in chromosomesList
+        if chromName and chromName.strip()
+    ]
+    if excludeChromsList:
+        chromosomesList = [
+            chromName
+            for chromName in chromosomesList
+            if chromName not in excludeChromsList
+        ]
+    if not chromosomesList:
+        raise ValueError(
+            "No valid chromosomes found after excluding specified chromosomes."
+        )
+
+    return core.genomeParams(
+        genomeName=genomeLabel,
+        chromSizesFile=chromSizesFile,
+        blacklistFile=blacklistFile,
+        sparseBedFile=sparseBedFile,
+        chromosomes=chromosomesList,
+        excludeChroms=excludeChromsList,
+        excludeForNorm=excludeForNormList,
+    )
+
+
+def getCountingArgs(config_path: str) -> core.countingParams:
+    configData = _loadConfig(config_path)
+
+    stepSize = _cfgGet(configData, "countingParams.stepSize", 25)
+    scaleDownFlag = _cfgGet(
+        configData, "countingParams.scaleDown", True
+    )
+    scaleFactorList = _cfgGet(
+        configData, "countingParams.scaleFactors", None
+    )
+    numReads = _cfgGet(configData, "countingParams.numReads", 100)
+    scaleFactorsControlList = _cfgGet(
+        configData, "countingParams.scaleFactorsControl", None
+    )
+    applyAsinhFlag = _cfgGet(
+        configData, "countingParams.applyAsinh", False
+    )
+    applyLogFlag = _cfgGet(
+        configData, "countingParams.applyLog", False
+    )
+
+    if applyAsinhFlag and applyLogFlag:
+        applyAsinhFlag = True
+        applyLogFlag = False
+        logger.warning(
+            "Both `applyAsinh` and `applyLog` are set. Overriding `applyLog` to False."
+        )
+
+    rescaleToTreatmentCoverageFlag = _cfgGet(
+        configData,
+        "countingParams.rescaleToTreatmentCoverage",
+        True,
+    )
+
+    if scaleFactorList is not None and not isinstance(
+        scaleFactorList, list
+    ):
+        raise ValueError("`scaleFactors` should be a list of floats.")
+
+    if scaleFactorsControlList is not None and not isinstance(
+        scaleFactorsControlList, list
+    ):
+        raise ValueError(
+            "`scaleFactorsControl` should be a list of floats."
+        )
+
+    if (
+        scaleFactorList is not None
+        and scaleFactorsControlList is not None
+        and len(scaleFactorList) != len(scaleFactorsControlList)
+    ):
+        if len(scaleFactorsControlList) == 1:
+            scaleFactorsControlList = scaleFactorsControlList * len(
+                scaleFactorList
+            )
+        else:
+            raise ValueError(
+                "control and treatment scale factors: must be equal length or 1 control"
+            )
+
+    return core.countingParams(
+        stepSize=stepSize,
+        scaleDown=scaleDownFlag,
+        scaleFactors=scaleFactorList,
+        scaleFactorsControl=scaleFactorsControlList,
+        numReads=numReads,
+        applyAsinh=applyAsinhFlag,
+        applyLog=applyLogFlag,
+        rescaleToTreatmentCoverage=rescaleToTreatmentCoverageFlag,
+    )
+
+
+def readConfig(config_path: str) -> Dict[str, Any]:
+    r"""Read and parse the configuration file for Consenrich.
+
+    :param config_path: Path to the YAML configuration file.
+    :return: Dictionary containing all parsed configuration parameters.
+    """
+    configData = _loadConfig(config_path)
+
+    inputParams = getInputArgs(config_path)
+    outputParams = getOutputArgs(config_path)
+    genomeParams = getGenomeArgs(config_path)
+    countingParams = getCountingArgs(config_path)
+
+    minRDefault = _getMinR(configData, len(inputParams.bamFiles))
+    minQDefault = (
+        minRDefault / len(inputParams.bamFiles)
+    ) + 0.10  # conditioning
+
+    matchingExcludeRegionsFileDefault: Optional[str] = (
+        genomeParams.blacklistFile
+    )
+
+    if (
+        inputParams.bamFilesControl is not None
+        and len(inputParams.bamFilesControl) > 0
+    ):
+        detrendWindowLengthBp = _cfgGet(
+            configData,
+            "detrendParams.detrendWindowLengthBP",
+            25_000,
+        )
+        detrendSavitzkyGolayDegree = _cfgGet(
+            configData,
+            "detrendParams.detrendSavitzkyGolayDegree",
+            1,
+        )
+    else:
+        detrendWindowLengthBp = _cfgGet(
+            configData,
+            "detrendParams.detrendWindowLengthBP",
+            10_000,
+        )
+        detrendSavitzkyGolayDegree = _cfgGet(
+            configData,
+            "detrendParams.detrendSavitzkyGolayDegree",
+            2,
+        )
+
+    experimentName = _cfgGet(
+        configData, "experimentName", "consenrichExperiment"
+    )
+
+    processArgs = core.processParams(
+        deltaF=_cfgGet(configData, "processParams.deltaF", 0.5),
+        minQ=_cfgGet(configData, "processParams.minQ", minQDefault),
+        maxQ=_cfgGet(configData, "processParams.maxQ", 500.0),
+        offDiagQ=_cfgGet(configData, "processParams.offDiagQ", 0.0),
+        dStatAlpha=_cfgGet(
+            configData, "processParams.dStatAlpha", 3.0
+        ),
+        dStatd=_cfgGet(configData, "processParams.dStatd", 10.0),
+        dStatPC=_cfgGet(configData, "processParams.dStatPC", 1.0),
+        scaleResidualsByP11=_cfgGet(
+            configData,
+            "processParams.scaleResidualsByP11",
+            False,
+        ),
+    )
+
+    observationArgs = core.observationParams(
+        minR=minRDefault,
+        maxR=_cfgGet(configData, "observationParams.maxR", 500.0),
+        useALV=_cfgGet(configData, "observationParams.useALV", False),
+        useConstantNoiseLevel=_cfgGet(
+            configData,
+            "observationParams.useConstantNoiseLevel",
+            False,
+        ),
+        noGlobal=_cfgGet(
+            configData, "observationParams.noGlobal", False
+        ),
+        numNearest=_cfgGet(
+            configData, "observationParams.numNearest", 25
+        ),
+        localWeight=_cfgGet(
+            configData, "observationParams.localWeight", 0.333
+        ),
+        globalWeight=_cfgGet(
+            configData, "observationParams.globalWeight", 0.667
+        ),
+        approximationWindowLengthBP=_cfgGet(
+            configData,
+            "observationParams.approximationWindowLengthBP",
+            10_000,
+        ),
+        lowPassWindowLengthBP=_cfgGet(
+            configData,
+            "observationParams.lowPassWindowLengthBP",
+            20_000,
+        ),
+        lowPassFilterType=_cfgGet(
+            configData,
+            "observationParams.lowPassFilterType",
+            "median",
+        ),
+        returnCenter=_cfgGet(
+            configData, "observationParams.returnCenter", True
+        ),
+    )
+
+    stateArgs = core.stateParams(
+        stateInit=_cfgGet(configData, "stateParams.stateInit", 0.0),
+        stateCovarInit=_cfgGet(
+            configData, "stateParams.stateCovarInit", 100.0
+        ),
+        boundState=_cfgGet(
+            configData, "stateParams.boundState", True
+        ),
+        stateLowerBound=_cfgGet(
+            configData, "stateParams.stateLowerBound", 0.0
+        ),
+        stateUpperBound=_cfgGet(
+            configData, "stateParams.stateUpperBound", 10000.0
+        ),
+    )
+
+    samThreads = _cfgGet(configData, "samParams.samThreads", 1)
+    samFlagExclude = _cfgGet(
+        configData, "samParams.samFlagExclude", 3844
+    )
+    oneReadPerBin = _cfgGet(configData, "samParams.oneReadPerBin", 0)
+    chunkSize = _cfgGet(configData, "samParams.chunkSize", 1_000_000)
+    offsetStr = _cfgGet(configData, "samParams.offsetStr", "0,0")
+    extendBpList = _cfgGet(configData, "samParams.extendBP", [])
+    maxInsertSize = _cfgGet(
+        configData, "samParams.maxInsertSize", 1000
+    )
+
+    pairedEndDefault = (
+        1
+        if inputParams.pairedEnd is not None
+        and int(inputParams.pairedEnd) > 0
+        else 0
+    )
+    inferFragmentDefault = (
+        1
+        if inputParams.pairedEnd is not None
+        and int(inputParams.pairedEnd) == 0
+        else 0
+    )
+
+    samArgs = core.samParams(
+        samThreads=samThreads,
+        samFlagExclude=samFlagExclude,
+        oneReadPerBin=oneReadPerBin,
+        chunkSize=chunkSize,
+        offsetStr=offsetStr,
+        extendBP=extendBpList,
+        maxInsertSize=maxInsertSize,
+        pairedEndMode=_cfgGet(
+            configData,
+            "samParams.pairedEndMode",
+            pairedEndDefault,
+        ),
+        inferFragmentLength=_cfgGet(
+            configData,
+            "samParams.inferFragmentLength",
+            inferFragmentDefault,
+        ),
+        countEndsOnly=_cfgGet(
+            configData, "samParams.countEndsOnly", False
+        ),
+    )
+
+    detrendArgs = core.detrendParams(
+        detrendWindowLengthBP=detrendWindowLengthBp,
+        detrendTrackPercentile=_cfgGet(
+            configData,
+            "detrendParams.detrendTrackPercentile",
+            75,
+        ),
+        usePolyFilter=_cfgGet(
+            configData, "detrendParams.usePolyFilter", False
+        ),
+        detrendSavitzkyGolayDegree=detrendSavitzkyGolayDegree,
+        useOrderStatFilter=_cfgGet(
+            configData, "detrendParams.useOrderStatFilter", True
+        ),
+    )
+
+    matchingArgs = core.matchingParams(
+        templateNames=_cfgGet(
+            configData, "matchingParams.templateNames", []
+        ),
+        cascadeLevels=_cfgGet(
+            configData, "matchingParams.cascadeLevels", []
+        ),
+        iters=_cfgGet(configData, "matchingParams.iters", 25_000),
+        alpha=_cfgGet(configData, "matchingParams.alpha", 0.05),
+        minMatchLengthBP=_cfgGet(
+            configData,
+            "matchingParams.minMatchLengthBP",
+            250,
+        ),
+        maxNumMatches=_cfgGet(
+            configData,
+            "matchingParams.maxNumMatches",
+            100_000,
+        ),
+        minSignalAtMaxima=_cfgGet(
+            configData,
+            "matchingParams.minSignalAtMaxima",
+            "q:0.75",
+        ),
+        merge=_cfgGet(configData, "matchingParams.merge", True),
+        mergeGapBP=_cfgGet(
+            configData, "matchingParams.mergeGapBP", None
+        ),
+        useScalingFunction=_cfgGet(
+            configData,
+            "matchingParams.useScalingFunction",
+            True,
+        ),
+        excludeRegionsBedFile=_cfgGet(
+            configData,
+            "matchingParams.excludeRegionsBedFile",
+            matchingExcludeRegionsFileDefault,
+        ),
+        randSeed=_cfgGet(configData, "matchingParams.randSeed", 42),
+        penalizeBy=_cfgGet(
+            configData, "matchingParams.penalizeBy", None
+        ),
+        eps=_cfgGet(configData, "matchingParams.eps", 1.0e-2),
+    )
+
+    return {
+        "experimentName": experimentName,
+        "genomeArgs": genomeParams,
+        "inputArgs": inputParams,
+        "outputArgs": outputParams,
+        "countingArgs": countingParams,
+        "processArgs": processArgs,
+        "observationArgs": observationArgs,
+        "stateArgs": stateArgs,
+        "samArgs": samArgs,
+        "detrendArgs": detrendArgs,
+        "matchingArgs": matchingArgs,
+    }
+
+
+def convertBedGraphToBigWig(experimentName, chromSizesFile,
+ suffixes: Optional[List[str]] = None):
+
+    if suffixes is None:
+        # at least look for `state` bedGraph
+        suffixes = ["state"]
+    path_ = ""
+    warningMessage = (
+        "Could not find UCSC bedGraphToBigWig binary utility."
+        "If you need bigWig files instead of the default, human-readable bedGraph files,"
+        "you can download the `bedGraphToBigWig` binary from https://hgdownload.soe.ucsc.edu/admin/exe/<operatingSystem, architecture>"
+        "OR install via conda (conda install -c bioconda ucsc-bedgraphtobigwig)."
+    )
+
+    logger.info(
+        "Attempting to generate bigWig files from bedGraph format..."
+    )
+    try:
+        path_ = shutil.which("bedGraphToBigWig")
+    except Exception as e:
+        logger.warning(f"\n{warningMessage}\n")
+        return
+    if path_ is None or len(path_) == 0:
+        logger.warning(f"\n{warningMessage}\n")
+        return
+    logger.info(f"Using bedGraphToBigWig from {path_}")
+    for suffix in suffixes:
+        bedgraph = (
+            f"consenrichOutput_{experimentName}_{suffix}.bedGraph"
+        )
+        if not os.path.exists(bedgraph):
+            logger.warning(
+                f"bedGraph file {bedgraph} does not exist. Skipping bigWig conversion."
+            )
+            continue
+        if not os.path.exists(chromSizesFile):
+            logger.warning(
+                f"{chromSizesFile} does not exist. Skipping bigWig conversion."
+            )
+            return
+        bigwig = f"{experimentName}_consenrich_{suffix}.bw"
+        logger.info(f"Start: {bedgraph} --> {bigwig}...")
+        try:
+            subprocess.run(
+                [path_, bedgraph, chromSizesFile, bigwig], check=True
+            )
+        except Exception as e:
+            logger.warning(
+                f"bedGraph-->bigWig conversion with\n\n\t`bedGraphToBigWig {bedgraph} {chromSizesFile} {bigwig}`\nraised: \n{e}\n\n"
+            )
+            continue
+        if os.path.exists(bigwig) and os.path.getsize(bigwig) > 100:
+            logger.info(
+                f"Finished: converted {bedgraph} to {bigwig}."
+            )
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Consenrich CLI")
+    parser.add_argument(
+        "--config",
+        type=str,
+        dest="config",
+        help="Path to a YAML config file with parameters + arguments defined in `consenrich.core`",
+    )
+
+    # --- Matching-specific command-line arguments ---
+    parser.add_argument(
+        "--match-bedGraph",
+        type=str,
+        dest="matchBedGraph",
+        help="Path to a bedGraph file of Consenrich estimates to match templates against.\
+            If provided, *only* the matching algorithm is run (no other processing). Note that \
+            some features in `consenrich.matching` may not be supported through this CLI interface.",
+    )
+    parser.add_argument(
+        "--match-template",
+        type=str,
+        default="haar",
+        choices=[
+            x
+            for x in pywt.wavelist(kind="discrete")
+            if "bio" not in x
+        ],
+        dest="matchTemplate",
+    )
+    parser.add_argument(
+        "--match-level", type=int, default=2, dest="matchLevel"
+    )
+    parser.add_argument(
+        "--match-alpha", type=float, default=0.05, dest="matchAlpha"
+    )
+    parser.add_argument(
+        "--match-min-length",
+        type=int,
+        default=250,
+        dest="matchMinMatchLengthBP",
+    )
+    parser.add_argument(
+        "--match-iters", type=int, default=25000, dest="matchIters"
+    )
+    parser.add_argument(
+        "--match-min-signal",
+        type=str,
+        default="q:0.75",
+        dest="matchMinSignalAtMaxima",
+    )
+    parser.add_argument(
+        "--match-max-matches",
+        type=int,
+        default=100000,
+        dest="matchMaxNumMatches",
+    )
+    parser.add_argument(
+        "--match-no-merge", action="store_true", dest="matchNoMerge"
+    )
+    parser.add_argument(
+        "--match-merge-gap",
+        type=int,
+        default=None,
+        dest="matchMergeGapBP",
+    )
+    parser.add_argument(
+        "--match-use-wavelet",
+        action="store_true",
+        dest="matchUseWavelet",
+    )
+    parser.add_argument(
+        "--match-seed", type=int, default=42, dest="matchRandSeed"
+    )
+    parser.add_argument(
+        "--match-exclude-bed",
+        type=str,
+        default=None,
+        dest="matchExcludeBed",
+    )
+    parser.add_argument(
+        "--verbose", action="store_true", help="If set, logs config"
+    )
+    args = parser.parse_args()
+
+    if args.matchBedGraph:
+        if not os.path.exists(args.matchBedGraph):
+            raise FileNotFoundError(
+                f"bedGraph file {args.matchBedGraph} couldn't be found."
+            )
+        logger.info(
+            f"Running matching algorithm using bedGraph file {args.matchBedGraph}..."
+        )
+
+        outName = matching.matchExistingBedGraph(
+            args.matchBedGraph,
+            args.matchTemplate,
+            args.matchLevel,
+            alpha=args.matchAlpha,
+            minMatchLengthBP=args.matchMinMatchLengthBP,
+            iters=args.matchIters,
+            minSignalAtMaxima=args.matchMinSignalAtMaxima,
+            maxNumMatches=args.matchMaxNumMatches,
+            useScalingFunction=(not args.matchUseWavelet),
+            merge=(not args.matchNoMerge),
+            mergeGapBP=args.matchMergeGapBP,
+            excludeRegionsBedFile=args.matchExcludeBed,
+            randSeed=args.matchRandSeed,
+        )
+        logger.info(f"Finished matching. Written to {outName}")
+        sys.exit(0)
+
+    if args.matchBedGraph:
+        # this shouldn't happen, but just in case -- matching on previous bedGraph means no other processing
+        logger.info(
+            "If `--match-bedgraph <path_to_bedgraph>` is provided, only the matching algorithm is run."
+        )
+        sys.exit(0)
+
+    if not args.config:
+        logger.info(
+            "No config file provided, run with `--config <path_to_config.yaml>`"
+        )
+        logger.info(
+            "See documentation: https://nolan-h-hamilton.github.io/Consenrich/"
+        )
+        sys.exit(1)
+
+    if not os.path.exists(args.config):
+        logger.info(f"Config file {args.config} does not exist.")
+        logger.info(
+            "See documentation: https://nolan-h-hamilton.github.io/Consenrich/"
+        )
+        sys.exit(1)
+
+    config = readConfig(args.config)
+    experimentName = config["experimentName"]
+    genomeArgs = config["genomeArgs"]
+    inputArgs = config["inputArgs"]
+    outputArgs = config["outputArgs"]
+    countingArgs = config["countingArgs"]
+    processArgs = config["processArgs"]
+    observationArgs = config["observationArgs"]
+    stateArgs = config["stateArgs"]
+    samArgs = config["samArgs"]
+    detrendArgs = config["detrendArgs"]
+    matchingArgs = config["matchingArgs"]
+    bamFiles = inputArgs.bamFiles
+    bamFilesControl = inputArgs.bamFilesControl
+    numSamples = len(bamFiles)
+    numNearest = observationArgs.numNearest
+    stepSize = countingArgs.stepSize
+    excludeForNorm = genomeArgs.excludeForNorm
+    chromSizes = genomeArgs.chromSizesFile
+    scaleDown = countingArgs.scaleDown
+    extendBP_ = core.resolveExtendBP(samArgs.extendBP, bamFiles)
+    initialTreatmentScaleFactors = []
+    minMatchLengthBP_: Optional[int] = matchingArgs.minMatchLengthBP
+    mergeGapBP_: Optional[int] = matchingArgs.mergeGapBP
+
+    if args.verbose:
+        try:
+            logger.info("Configuration:\n")
+            config_truncated = {
+                k: v
+                for k, v in config.items()
+                if k
+                not in ["inputArgs", "genomeArgs", "countingArgs"]
+            }
+            config_truncated["experimentName"] = experimentName
+            config_truncated["inputArgs"] = inputArgs
+            config_truncated["outputArgs"] = outputArgs
+            config_truncated["genomeArgs"] = genomeArgs
+            config_truncated["countingArgs"] = countingArgs
+            config_truncated["processArgs"] = processArgs
+            config_truncated["observationArgs"] = observationArgs
+            config_truncated["stateArgs"] = stateArgs
+            config_truncated["samArgs"] = samArgs
+            config_truncated["detrendArgs"] = detrendArgs
+            pprint.pprint(config_truncated, indent=8)
+        except Exception as e:
+            logger.warning(f"Failed to print parsed config:\n{e}\n")
+
+    controlsPresent = checkControlsPresent(inputArgs)
+    if args.verbose:
+        logger.info(f"controlsPresent: {controlsPresent}")
+    readLengthsBamFiles = getReadLengths(
+        inputArgs, countingArgs, samArgs
+    )
+    effectiveGenomeSizes = getEffectiveGenomeSizes(
+        genomeArgs, readLengthsBamFiles
+    )
+    matchingEnabled = checkMatchingEnabled(matchingArgs)
+    if args.verbose:
+        logger.info(f"matchingEnabled: {matchingEnabled}")
+    scaleFactors = countingArgs.scaleFactors
+    scaleFactorsControl = countingArgs.scaleFactorsControl
+
+    if controlsPresent:
+        readLengthsControlBamFiles = [
+            core.getReadLength(
+                bamFile,
+                countingArgs.numReads,
+                1000,
+                samArgs.samThreads,
+                samArgs.samFlagExclude,
+            )
+            for bamFile in bamFilesControl
+        ]
+        effectiveGenomeSizesControl = [
+            constants.getEffectiveGenomeSize(
+                genomeArgs.genomeName, readLength
+            )
+            for readLength in readLengthsControlBamFiles
+        ]
+
+        if (
+            scaleFactors is not None
+            and scaleFactorsControl is not None
+        ):
+            treatScaleFactors = scaleFactors
+            controlScaleFactors = scaleFactorsControl
+            # still make sure this is accessible
+            initialTreatmentScaleFactors = [1.0] * len(bamFiles)
+        else:
+            try:
+                initialTreatmentScaleFactors = [
+                    detrorm.getScaleFactor1x(
+                        bamFile,
+                        effectiveGenomeSize,
+                        readLength,
+                        genomeArgs.excludeChroms,
+                        genomeArgs.chromSizesFile,
+                        samArgs.samThreads,
+                    )
+                    for bamFile, effectiveGenomeSize, readLength in zip(
+                        bamFiles,
+                        effectiveGenomeSizes,
+                        readLengthsBamFiles,
+                    )
+                ]
+            except Exception:
+                initialTreatmentScaleFactors = [1.0] * len(bamFiles)
+
+            pairScalingFactors = [
+                detrorm.getPairScaleFactors(
+                    bamFileA,
+                    bamFileB,
+                    effectiveGenomeSizeA,
+                    effectiveGenomeSizeB,
+                    readLengthA,
+                    readLengthB,
+                    excludeForNorm,
+                    chromSizes,
+                    samArgs.samThreads,
+                    scaleDown,
+                )
+                for bamFileA, bamFileB, effectiveGenomeSizeA, effectiveGenomeSizeB, readLengthA, readLengthB in zip(
+                    bamFiles,
+                    bamFilesControl,
+                    effectiveGenomeSizes,
+                    effectiveGenomeSizesControl,
+                    readLengthsBamFiles,
+                    readLengthsControlBamFiles,
+                )
+            ]
+
+            treatScaleFactors = []
+            controlScaleFactors = []
+            for scaleFactorA, scaleFactorB in pairScalingFactors:
+                treatScaleFactors.append(scaleFactorA)
+                controlScaleFactors.append(scaleFactorB)
+
+    else:
+        treatScaleFactors = scaleFactors
+        controlScaleFactors = scaleFactorsControl
+
+    if scaleFactors is None and not controlsPresent:
+        scaleFactors = [
+            detrorm.getScaleFactor1x(
+                bamFile,
+                effectiveGenomeSize,
+                readLength,
+                genomeArgs.excludeChroms,
+                genomeArgs.chromSizesFile,
+                samArgs.samThreads,
+            )
+            for bamFile, effectiveGenomeSize, readLength in zip(
+                bamFiles, effectiveGenomeSizes, readLengthsBamFiles
+            )
+        ]
+    chromSizesDict = misc_util.getChromSizesDict(
+        genomeArgs.chromSizesFile,
+        excludeChroms=genomeArgs.excludeChroms,
+    )
+    chromosomes = genomeArgs.chromosomes
+
+    for c_, chromosome in enumerate(chromosomes):
+        chromosomeStart, chromosomeEnd = core.getChromRangesJoint(
+            bamFiles,
+            chromosome,
+            chromSizesDict[chromosome],
+            samArgs.samThreads,
+            samArgs.samFlagExclude,
+        )
+        chromosomeStart = max(
+            0, (chromosomeStart - (chromosomeStart % stepSize))
+        )
+        chromosomeEnd = max(
+            0, (chromosomeEnd - (chromosomeEnd % stepSize))
+        )
+        numIntervals = (
+            ((chromosomeEnd - chromosomeStart) + stepSize) - 1
+        ) // stepSize
+        intervals = np.arange(
+            chromosomeStart, chromosomeEnd, stepSize
+        )
+        chromMat: np.ndarray = np.empty(
+            (numSamples, numIntervals), dtype=np.float32
+        )
+        if controlsPresent:
+            j_: int = 0
+            finalSF = 1.0
+            for bamA, bamB in zip(bamFiles, bamFilesControl):
+                logger.info(
+                    f"Counting (trt,ctrl) for {chromosome}: ({bamA}, {bamB})"
+                )
+                pairMatrix: np.ndarray = core.readBamSegments(
+                    [bamA, bamB],
+                    chromosome,
+                    chromosomeStart,
+                    chromosomeEnd,
+                    stepSize,
+                    [
+                        readLengthsBamFiles[j_],
+                        readLengthsControlBamFiles[j_],
+                    ],
+                    [treatScaleFactors[j_], controlScaleFactors[j_]],
+                    samArgs.oneReadPerBin,
+                    samArgs.samThreads,
+                    samArgs.samFlagExclude,
+                    offsetStr=samArgs.offsetStr,
+                    extendBP=extendBP_[j_],
+                    maxInsertSize=samArgs.maxInsertSize,
+                    pairedEndMode=samArgs.pairedEndMode,
+                    inferFragmentLength=samArgs.inferFragmentLength,
+                    applyAsinh=countingArgs.applyAsinh,
+                    applyLog=countingArgs.applyLog,
+                    countEndsOnly=samArgs.countEndsOnly,
+                )
+                if countingArgs.rescaleToTreatmentCoverage:
+                    finalSF = max(
+                        1.0, initialTreatmentScaleFactors[j_]
+                    )
+                chromMat[j_, :] = finalSF * (
+                    pairMatrix[0, :] - pairMatrix[1, :]
+                )
+                j_ += 1
+        else:
+            chromMat = core.readBamSegments(
+                bamFiles,
+                chromosome,
+                chromosomeStart,
+                chromosomeEnd,
+                stepSize,
+                readLengthsBamFiles,
+                scaleFactors,
+                samArgs.oneReadPerBin,
+                samArgs.samThreads,
+                samArgs.samFlagExclude,
+                offsetStr=samArgs.offsetStr,
+                extendBP=extendBP_,
+                maxInsertSize=samArgs.maxInsertSize,
+                pairedEndMode=samArgs.pairedEndMode,
+                inferFragmentLength=samArgs.inferFragmentLength,
+                applyAsinh=countingArgs.applyAsinh,
+                applyLog=countingArgs.applyLog,
+                countEndsOnly=samArgs.countEndsOnly,
+            )
+        sparseMap = None
+        if genomeArgs.sparseBedFile and not observationArgs.useALV:
+            logger.info(
+                f"Building sparse mapping for {chromosome}..."
+            )
+            sparseMap = core.getSparseMap(
+                chromosome,
+                intervals,
+                numNearest,
+                genomeArgs.sparseBedFile,
+            )
+
+        muncMat = np.empty_like(chromMat, dtype=np.float32)
+        for j in range(numSamples):
+            logger.info(
+                f"Muncing {j + 1}/{numSamples} for {chromosome}..."
+            )
+            muncMat[j, :] = core.getMuncTrack(
+                chromosome,
+                intervals,
+                stepSize,
+                chromMat[j, :],
+                observationArgs.minR,
+                observationArgs.maxR,
+                observationArgs.useALV,
+                observationArgs.useConstantNoiseLevel,
+                observationArgs.noGlobal,
+                observationArgs.localWeight,
+                observationArgs.globalWeight,
+                observationArgs.approximationWindowLengthBP,
+                observationArgs.lowPassWindowLengthBP,
+                observationArgs.returnCenter,
+                sparseMap=sparseMap,
+                lowPassFilterType=observationArgs.lowPassFilterType,
+            )
+            chromMat[j, :] = detrorm.detrendTrack(
+                chromMat[j, :],
+                stepSize,
+                detrendArgs.detrendWindowLengthBP,
+                detrendArgs.useOrderStatFilter,
+                detrendArgs.usePolyFilter,
+                detrendArgs.detrendTrackPercentile,
+                detrendArgs.detrendSavitzkyGolayDegree,
+            )
+        logger.info(f">>>Running consenrich: {chromosome}<<<")
+
+        x, P, y = core.runConsenrich(
+            chromMat,
+            muncMat,
+            processArgs.deltaF,
+            processArgs.minQ,
+            processArgs.maxQ,
+            processArgs.offDiagQ,
+            processArgs.dStatAlpha,
+            processArgs.dStatd,
+            processArgs.dStatPC,
+            stateArgs.stateInit,
+            stateArgs.stateCovarInit,
+            stateArgs.boundState,
+            stateArgs.stateLowerBound,
+            stateArgs.stateUpperBound,
+            samArgs.chunkSize,
+            progressIter=50_000,
+        )
+        logger.info("Done.")
+
+        x_ = core.getPrimaryState(x)
+        y_ = core.getPrecisionWeightedResidual(
+            y,
+            muncMat,
+            stateCovarSmoothed=P
+            if processArgs.scaleResidualsByP11 is not None
+            and processArgs.scaleResidualsByP11
+            else None,
+        )
+
+        weights_: Optional[np.ndarray] = None
+        if matchingArgs.penalizeBy is not None:
+            if matchingArgs.penalizeBy == "absResiduals":
+                try:
+                    weights_ = np.abs(y_)
+                except Exception as e:
+                    logger.warning(
+                        f"Error computing weights for 'absResiduals': {e}. No weights applied for matching."
+                    )
+                    weights_ = None
+            elif matchingArgs.penalizeBy == "stateUncertainty" or matchingArgs.penalizeBy == "stateStdDev":
+                try:
+                    weights_ = np.sqrt(P[:, 0, 0])
+                except Exception as e:
+                    logger.warning(
+                        f"Error computing weights for 'stateUncertainty': {e}. No weights applied for matching."
+                    )
+                    weights_ = None
+            elif matchingArgs.penalizeBy == "muncTrace":
+                try:
+                    weights_ = np.sqrt(
+                        np.trace(muncMat, axis1=0, axis2=1)
+                        / numSamples
+                    )
+                except Exception as e:
+                    logger.warning(
+                        f"Error computing weights for 'muncTrace': {e}. No weights applied for matching."
+                    )
+                    weights_ = None
+            else:
+                logger.warning(
+                    f"Unrecognized `matchingParams.penalizeBy`: {matchingArgs.penalizeBy}. No weights applied."
+                )
+                weights_ = None
+
+        df = pd.DataFrame(
+            {
+                "Chromosome": chromosome,
+                "Start": intervals,
+                "End": intervals + stepSize,
+                "State": x_,
+            }
+        )
+
+        if outputArgs.writeResiduals:
+            df["Res"] = y_.astype(np.float32) # FFR: cast necessary?
+        if outputArgs.writeMuncTrace:
+            df["Munc"] = np.sqrt(np.trace(muncMat, axis1=0, axis2=1) / numSamples).astype(np.float32)
+        if outputArgs.writeStateStd:
+            df["StateStd"] = np.sqrt(P[:, 0, 0]).astype(np.float32)
+        cols_ = ["Chromosome", "Start", "End", "State"]
+        if outputArgs.writeResiduals:
+            cols_.append("Res")
+        if outputArgs.writeMuncTrace:
+            cols_.append("Munc")
+        if outputArgs.writeStateStd:
+            cols_.append("StateStd")
+        df = df[cols_]
+        suffixes = ['state']
+        if outputArgs.writeResiduals:
+            suffixes.append('residuals')
+        if outputArgs.writeMuncTrace:
+            suffixes.append('muncTraces')
+        if outputArgs.writeStateStd:
+            suffixes.append('stdDevs')
+
+        if c_ == 0 and len(chromosomes) > 1:
+            for file_ in os.listdir("."):
+                if file_.startswith(
+                    f"consenrichOutput_{experimentName}"
+                ) and (
+                    file_.endswith(".bedGraph")
+                    or file_.endswith(".narrowPeak")
+                ):
+                    logger.warning(f"Overwriting: {file_}")
+                    os.remove(file_)
+
+        for col, suffix in zip(cols_[3:], suffixes):
+            logger.info(
+                f"{chromosome}: writing/appending to: consenrichOutput_{experimentName}_{suffix}.bedGraph"
+            )
+            df[["Chromosome", "Start", "End", col]].to_csv(
+                f"consenrichOutput_{experimentName}_{suffix}.bedGraph",
+                sep="\t",
+                header=False,
+                index=False,
+                mode="a",
+                float_format="%.3f",
+                lineterminator="\n",
+            )
+        try:
+            if matchingEnabled:
+                if (
+                    minMatchLengthBP_ is None
+                    or minMatchLengthBP_ <= 0
+                ):
+                    minMatchLengthBP_ = (
+                        matching.autoMinLengthIntervals(x_)
+                        * (intervals[1] - intervals[0])
+                    )
+
+                if mergeGapBP_ is None:
+                    mergeGapBP_ = int(minMatchLengthBP_ / 2) + 1
+
+                matchingDF = matching.matchWavelet(
+                    chromosome,
+                    intervals,
+                    x_,
+                    matchingArgs.templateNames,
+                    matchingArgs.cascadeLevels,
+                    matchingArgs.iters,
+                    matchingArgs.alpha,
+                    minMatchLengthBP_,
+                    matchingArgs.maxNumMatches,
+                    matchingArgs.minSignalAtMaxima,
+                    useScalingFunction=matchingArgs.useScalingFunction,
+                    excludeRegionsBedFile=matchingArgs.excludeRegionsBedFile,
+                    randSeed=matchingArgs.randSeed,
+                    weights=weights_,
+                )
+                if not matchingDF.empty:
+                    matchingDF.to_csv(
+                        f"consenrichOutput_{experimentName}_matches.narrowPeak",
+                        sep="\t",
+                        header=False,
+                        index=False,
+                        mode="a",
+                        float_format=f"%.{outputArgs.roundDigits}f",
+                        lineterminator="\n",
+                    )
+        except Exception as e:
+            logger.warning(
+                f"Matching routine unsuccessful for {chromosome}...SKIPPING:\n{e}\n\n"
+            )
+            continue
+    logger.info("Finished: output in human-readable format")
+
+    if outputArgs.convertToBigWig:
+        convertBedGraphToBigWig(experimentName, genomeArgs.chromSizesFile, suffixes=suffixes)
+
+    if matchingEnabled and matchingArgs.merge:
+        try:
+            mergeGapBP_ = matchingArgs.mergeGapBP
+            if mergeGapBP_ is None or mergeGapBP_ <= 0:
+                mergeGapBP_ = (
+                    int(minMatchLengthBP_ / 2) + 1
+                    if minMatchLengthBP_ is not None
+                    and minMatchLengthBP_ >= 0
+                    else 75
+                )
+            matching.mergeMatches(
+                f"consenrichOutput_{experimentName}_matches.narrowPeak",
+                mergeGapBP=mergeGapBP_,
+            )
+
+        except Exception as e:
+            logger.warning(
+                f"Failed to merge matches...SKIPPING:\n{e}\n\n"
+            )
+    logger.info("Done.")
+
+
+if __name__ == "__main__":
+    main()