Source code for cnvlib.fix

"""Supporting functions for the 'fix' command."""

from __future__ import annotations
import logging
from typing import TYPE_CHECKING, Optional, Union

import numpy as np
import pandas as pd

from . import descriptives, params, smoothing

if TYPE_CHECKING:
    from cnvlib.cnary import CopyNumArray
    from numpy import ndarray
    from pandas.core.series import Series


[docs] def do_fix( target_raw: CopyNumArray, antitarget_raw: CopyNumArray, reference: CopyNumArray, diploid_parx_genome: Optional[str] = None, do_gc: bool = True, do_edge: bool = True, do_rmask: bool = True, do_cluster: bool = False, smoothing_window_fraction: None = None, ) -> CopyNumArray: """Normalize tumor/test sample coverage using a reference and correct biases. This is the core normalization step in the CNVkit pipeline. It combines target and antitarget coverage data, applies the reference baseline, corrects for systematic biases (GC content, edge effects, RepeatMasker), and produces log2 copy ratios ready for segmentation. The "fix" name refers to fixing/correcting coverage biases. Parameters ---------- target_raw : CopyNumArray Raw coverage data for on-target/baited regions from the test sample. Typically a .targetcoverage.cnn file from the `coverage` command. antitarget_raw : CopyNumArray Raw coverage data for off-target/antitarget regions from the test sample. Typically a .antitargetcoverage.cnn file from the `coverage` command. Can be empty for amplicon sequencing. reference : CopyNumArray Pooled reference created from normal samples using `do_reference`. Contains expected neutral coverage and bias correction features. diploid_parx_genome : str, optional Reference genome name for pseudo-autosomal region handling (e.g., 'hg19', 'hg38', 'mm10'). Ensures PAR regions on X/Y chromosomes are treated as diploid during normalization. Default: None do_gc : bool, optional Apply GC bias correction using GC content from reference. Corrects for systematic coverage variations due to GC%. Default: True do_edge : bool, optional Apply edge effect correction for target bins. Corrects for reduced coverage near bait interval boundaries. Default: True do_rmask : bool, optional Apply RepeatMasker correction for antitarget bins. Corrects for coverage variations in repetitive regions. Default: True do_cluster : bool, optional If the reference contains multiple sub-clusters (from hierarchical clustering during reference creation), select the cluster with highest correlation to this sample. Default: False smoothing_window_fraction : None, optional Reserved parameter for future smoothing functionality. Currently not implemented. Default: None Returns ------- CopyNumArray Normalized copy number ratios (.cnr file) with log2 ratios centered around 0 (neutral copy number). Ready for segmentation with the `segment` command. Notes ----- The normalization process: 1. **Bias correction** (separately for targets and antitargets): - GC correction: Adjusts for GC content-dependent coverage bias - Edge correction: Adjusts for lower coverage near bait edges (targets only) - RepeatMasker correction: Adjusts for repeats (antitargets only) 2. **Reference subtraction**: - Subtracts reference log2 values from sample log2 values - If do_cluster=True, selects best-matching reference cluster 3. **Weighting**: - Applies statistical weights based on coverage variability - Down-weights unreliable bins 4. **Centering**: - Centers the genome-wide distribution around log2=0 - Accounts for sex chromosomes and PAR regions The output .cnr file contains one row per bin with columns: - chromosome, start, end, gene: Genomic location - log2: Copy ratio in log2 scale (0 = diploid, +1 = gain, -1 = loss) - depth: Read depth - weight: Statistical weight for segmentation See Also -------- do_reference : Creates the reference used for normalization load_adjust_coverages : Internal function that performs bias correction apply_weights : Calculates statistical weights for bins Examples -------- Basic usage: >>> fixed = do_fix( ... target_raw=read_cna('Sample.targetcoverage.cnn'), ... antitarget_raw=read_cna('Sample.antitargetcoverage.cnn'), ... reference=read_cna('reference.cnn') ... ) With cluster selection: >>> fixed = do_fix( ... target_raw, antitarget_raw, reference, ... do_cluster=True # Select best-matching reference cluster ... ) For male sample with PAR handling: >>> fixed = do_fix( ... target_raw, antitarget_raw, reference, ... diploid_parx_genome='hg38' ... ) """ # Load, recenter and GC-correct target & antitarget probes separately logging.info("Processing target: %s", target_raw.sample_id) cnarr, ref_matched = load_adjust_coverages( target_raw, reference, True, do_gc, do_edge, False, diploid_parx_genome, smoothing_window_fraction=smoothing_window_fraction, ) logging.info("Processing antitarget: %s", antitarget_raw.sample_id) anti_cnarr, ref_anti = load_adjust_coverages( antitarget_raw, reference, False, do_gc, False, do_rmask, diploid_parx_genome, smoothing_window_fraction=smoothing_window_fraction, ) if len(anti_cnarr): # Combine target and antitarget bins cnarr.add(anti_cnarr) ref_matched.add(ref_anti) # Find reference clusters, if requested log2_key = "log2" spread_key = "spread" if do_cluster: ref_log2_cols = [ col for col in ref_matched.data.columns if col == "log2" or col.startswith("log2") ] if len(ref_log2_cols) == 1: logging.info( "Reference does not contain any sub-clusters; using %s", log2_key ) else: # Get correlations between test sample and each reference cluster corr_coefs = np.array( [cnarr.log2.corr(ref_matched[ref_col]) for ref_col in ref_log2_cols] ) ordered = [ (k, r) for r, k in sorted(zip(corr_coefs, ref_log2_cols, strict=False), reverse=True) ] logging.info( "Correlations with each cluster:\n\t%s", "\n\t".join([f"{k}\t: {r}" for k, r in ordered]), ) log2_key = ordered[0][0] if log2_key.startswith("log2_"): suffix = log2_key.split("_", 1)[1] spread_key = "spread_" + suffix logging.info(" -> Choosing columns %r and %r", log2_key, spread_key) # Normalize coverages according to the reference # (Subtract the reference log2 copy number to get the log2 ratio) cnarr.data["log2"] -= ref_matched[log2_key] cnarr = apply_weights(cnarr, ref_matched, log2_key, spread_key) cnarr.center_all(skip_low=True, diploid_parx_genome=diploid_parx_genome) return cnarr
[docs] def load_adjust_coverages( cnarr: CopyNumArray, ref_cnarr: CopyNumArray, skip_low: bool, fix_gc: bool, fix_edge: bool, fix_rmask: bool, diploid_parx_genome: Optional[str], smoothing_window_fraction: None = None, ) -> tuple[CopyNumArray, CopyNumArray]: """Load and filter probe coverages; correct using reference and GC.""" if "gc" in cnarr: # Don't choke on Picard-derived files that have the GC column cnarr = cnarr.keep_columns((*cnarr._required_columns, "depth")) # No corrections needed if there are no data rows (e.g. no antitargets) if not len(cnarr): return cnarr, ref_cnarr[:0] ref_matched = match_ref_to_sample(ref_cnarr, cnarr) # Drop bins that had poor coverage in the pooled reference ok_cvg_indices = ~mask_bad_bins(ref_matched) logging.info("Keeping %d of %d bins", sum(ok_cvg_indices), len(ref_matched)) cnarr = cnarr[ok_cvg_indices] ref_matched = ref_matched[ok_cvg_indices] # Apply corrections for known systematic biases in coverage cnarr.center_all(skip_low=skip_low, diploid_parx_genome=diploid_parx_genome) # Skip bias corrections if most bins have no coverage (e.g. user error) if ( cnarr["log2"] > params.NULL_LOG2_COVERAGE - params.MIN_REF_COVERAGE ).sum() <= len(cnarr) // 2: logging.warning( "WARNING: most bins have no or very low coverage; " "check that the right BED file was used" ) else: # Smoothing window fraction converges on percentile binning, like Picard frac = smoothing_window_fraction if frac is None: frac = max(0.01, len(cnarr) ** -0.5) cnarr_index_reset = False if fix_gc: if "gc" in ref_matched: logging.info("Correcting for GC bias...") cnarr = center_by_window(cnarr, frac, ref_matched["gc"]) cnarr_index_reset = True else: logging.warning("WARNING: Skipping correction for GC bias") if fix_edge: logging.info("Correcting for density bias...") edge_bias = get_edge_bias(cnarr, params.INSERT_SIZE) cnarr = center_by_window(cnarr, frac, edge_bias) cnarr_index_reset = True if fix_rmask: if "rmask" in ref_matched: logging.info("Correcting for RepeatMasker bias...") cnarr = center_by_window(cnarr, frac, ref_matched["rmask"]) cnarr_index_reset = True else: logging.warning("WARNING: Skipping correction for RepeatMasker bias") if cnarr_index_reset: ref_matched.data = ref_matched.data.reset_index(drop=True) return cnarr, ref_matched
[docs] def mask_bad_bins(cnarr: CopyNumArray) -> Series: """Flag the bins with excessively low or inconsistent coverage. Returns ------- np.array A boolean array where True indicates bins that failed the checks. """ mask = ( (cnarr["log2"] < params.MIN_REF_COVERAGE) | (cnarr["log2"] > -params.MIN_REF_COVERAGE) | (cnarr["spread"] > params.MAX_REF_SPREAD) ) if "depth" in cnarr: mask |= cnarr["depth"] == 0 if "gc" in cnarr: assert params.GC_MIN_FRACTION >= 0 and params.GC_MIN_FRACTION <= 1 assert params.GC_MAX_FRACTION >= 0 and params.GC_MAX_FRACTION <= 1 lower_gc_bound = min(params.GC_MIN_FRACTION, params.GC_MAX_FRACTION) upper_gc_bound = max(params.GC_MIN_FRACTION, params.GC_MAX_FRACTION) mask |= (cnarr["gc"] > upper_gc_bound) | (cnarr["gc"] < lower_gc_bound) return mask
[docs] def match_ref_to_sample( ref_cnarr: CopyNumArray, samp_cnarr: CopyNumArray ) -> CopyNumArray: """Filter the reference bins to match the sample (target or antitarget).""" # Assign each bin a unique string ID based on genomic coordinates samp_labeled = samp_cnarr.data.set_index(pd.Index(samp_cnarr.coords())) ref_labeled = ref_cnarr.data.set_index(pd.Index(ref_cnarr.coords())) for dset, name in ((samp_labeled, "sample"), (ref_labeled, "reference")): dupes = dset.index.duplicated() if dupes.any(): raise ValueError( ( "Duplicated genomic coordinates in {} set. Total duplicated regions: {}, starting with:\n" "{}." ).format( name, len(dset.index[dupes]), "\n".join(map(str, dset.index[dupes][:10])), ) ) # Take the reference bins with IDs identical to those in the sample ref_matched = ref_labeled.reindex(index=samp_labeled.index) # Check for signs that the wrong reference was used num_missing = pd.isna(ref_matched.start).sum() if num_missing > 0: raise ValueError( f"Reference is missing {num_missing} bins found in {samp_cnarr.sample_id}" ) x = ref_cnarr.as_dataframe( ref_matched.reset_index(drop=True).set_index(samp_cnarr.data.index) ) return x
[docs] def center_by_window( cnarr: CopyNumArray, fraction: float, sort_key: Union[Series, ndarray] ) -> CopyNumArray: """Smooth out biases according to the trait specified by sort_key. E.g. correct GC-biased bins by windowed averaging across similar-GC bins; or for similar interval sizes. """ # Separate neighboring bins that could have the same key # (to avoid re-centering actual CNV regions -- only want an independently # sampled subset of presumably overall-CN-neutral bins) df = cnarr.data.reset_index(drop=True) rng = np.random.default_rng(0xA5EED) shuffle_order = rng.permutation(df.index) # df = df.reindex(shuffle_order) df = df.iloc[shuffle_order] # Apply the same shuffling to the key array as to the target probe set if isinstance(sort_key, pd.Series): # XXX why sort_key = sort_key.to_numpy() sort_key = sort_key[shuffle_order] # Sort the data according to the specified parameter order = np.argsort(sort_key, kind="mergesort") df = df.iloc[order] biases = smoothing.rolling_median(df["log2"], fraction) # biases = smoothing.savgol(df['log2'], fraction) df["log2"] -= biases fixarr = cnarr.as_dataframe(df) fixarr.sort() return fixarr
[docs] def get_edge_bias(cnarr: CopyNumArray, margin: int) -> Series: """Quantify the "edge effect" of the target tile and its neighbors. The result is proportional to the change in the target's coverage due to these edge effects, i.e. the expected loss of coverage near the target edges and, if there are close neighboring tiles, gain of coverage due to "spill over" reads from the neighbor tiles. (This is not the actual change in coverage. This is just a tribute.) """ output_by_chrom = [] for _chrom, subarr in cnarr.by_chromosome(): tile_starts = subarr["start"].to_numpy() tile_ends = subarr["end"].to_numpy() tgt_sizes = tile_ends - tile_starts # Calculate coverage loss at (both edges of) each tile losses = edge_losses(tgt_sizes, margin) # Find tiled intervals within a margin (+/- bp) of the given probe # (excluding the probe itself), then calculate the relative coverage # "gain" due to the neighbors, if any gap_sizes = tile_starts[1:] - tile_ends[:-1] ok_gaps_mask = gap_sizes < margin ok_gaps = gap_sizes[ok_gaps_mask] left_gains = edge_gains(tgt_sizes[1:][ok_gaps_mask], ok_gaps, margin) right_gains = edge_gains(tgt_sizes[:-1][ok_gaps_mask], ok_gaps, margin) gains = np.zeros(len(subarr)) gains[np.concatenate([[False], ok_gaps_mask])] += left_gains gains[np.concatenate([ok_gaps_mask, [False]])] += right_gains output_by_chrom.append(gains - losses) return pd.Series(np.concatenate(output_by_chrom), index=cnarr.data.index)
[docs] def edge_losses(target_sizes: ndarray, insert_size: int) -> ndarray: """Calculate coverage losses at the edges of baited regions. Letting i = insert size and t = target size, the proportional loss of coverage near the two edges of the baited region (combined) is: .. math :: i/2t If the "shoulders" extend outside the bait $(t < i), reduce by: .. math :: (i-t)^2 / 4it on each side, or (i-t)^2 / 2it total. """ losses = insert_size / (2 * target_sizes) # Drop the shoulder part that would extend past the bait small_mask = target_sizes < insert_size t_small = target_sizes[small_mask] losses[small_mask] -= (insert_size - t_small) ** 2 / (2 * insert_size * t_small) return losses
[docs] def edge_gains(target_sizes: ndarray, gap_sizes: ndarray, insert_size: int) -> ndarray: """Calculate coverage gain from neighboring baits' flanking reads. Letting i = insert size, t = target size, g = gap to neighboring bait, the gain of coverage due to a nearby bait, if g < i, is:: .. math :: (i-g)^2 / 4it If the neighbor flank extends beyond the target (t+g < i), reduce by:: .. math :: (i-t-g)^2 / 4it If a neighbor overlaps the target, treat it as adjacent (gap size 0). """ if not (gap_sizes <= insert_size).all(): raise ValueError( "Gaps greater than insert size:\n" + gap_sizes[gap_sizes > insert_size].head() ) gap_sizes = np.maximum(0, gap_sizes) gains = (insert_size - gap_sizes) ** 2 / (4 * insert_size * target_sizes) # Drop the flank part that extends past this baited region past_other_side_mask = target_sizes + gap_sizes < insert_size g_past = gap_sizes[past_other_side_mask] t_past = target_sizes[past_other_side_mask] gains[past_other_side_mask] -= (insert_size - t_past - g_past) ** 2 / ( 4 * insert_size * t_past ) return gains
[docs] def apply_weights( cnarr: CopyNumArray, ref_matched: CopyNumArray, log2_key: str, spread_key: str, epsilon: float = 1e-4, ) -> CopyNumArray: """Calculate weights for each bin. Bin weight is an estimate of (1 - variance) and within the range ``(0, 1]``. Weights are derived from: - Each bin's size - Sample's genome-wide average (on/off-target) coverage depth - Sample's genome-wide observed (on/off-target) bin variances And with a pooled reference: - Each bin's coverage depth in the reference - The "spread" column of the reference (approx. stdev) These estimates of variance assume the number of aligned reads per bin follows a Poisson distribution, approximately log-normal. Parameters ---------- cnarr : CopyNumArray Sample bins. ref_match : CopyNumArray Reference bins. log2_key : string The 'log2' column name in the reference to use. A clustered reference may have a suffix indicating the cluster, e.g. "log2_1". spread_key : string The 'spread' or 'spread_<cluster_id>' column name to use. epsilon : float Minimum value for bin weights, to avoid 0-weight bins causing errors later during segmentation. (CBS doesn't allow 0-weight bins.) Returns: The input `cnarr` with a `weight` column added. """ # Weight by sample-level features -- works for flat reference, too logging.debug("Weighting bins by size and overall variance in sample") simple_wt = np.zeros(len(cnarr)) # Calculate separately for on-, off-target bins is_anti = cnarr["gene"].isin(params.ANTITARGET_ALIASES) tgt_cna = cnarr[~is_anti] tgt_var = ( descriptives.biweight_midvariance(tgt_cna.drop_low_coverage().residuals()) ** 2 ) bin_sz = np.sqrt(tgt_cna["end"] - tgt_cna["start"]) tgt_simple_wts = 1 - tgt_var / (bin_sz / bin_sz.mean()) simple_wt[~is_anti] = tgt_simple_wts if is_anti.any(): # Check for a common user error anti_cna = cnarr[is_anti] anti_ok = anti_cna.drop_low_coverage() frac_anti_low = 1 - (len(anti_ok) / len(anti_cna)) if frac_anti_low > 0.5: # Off-target bins are mostly garbage -- skip reweighting logging.warning( "WARNING: Most antitarget bins ({:.2f}%, {:d}/{:d})" " have low or no coverage; is this amplicon/WGS?".format( 100 * frac_anti_low, len(anti_cna) - len(anti_ok), len(anti_cna) ) ) anti_var = descriptives.biweight_midvariance(anti_ok.residuals()) ** 2 anti_bin_sz = np.sqrt(anti_cna["end"] - anti_cna["start"]) anti_simple_wts = 1 - anti_var / (anti_bin_sz / anti_bin_sz.mean()) simple_wt[is_anti] = anti_simple_wts # Report any difference in bin set variability var_ratio = max(tgt_var, 0.01) / max(anti_var, 0.01) if var_ratio > 1: logging.info("Targets are %.2f x more variable than antitargets", var_ratio) else: logging.info( "Antitargets are %.2f x more variable than targets", 1.0 / var_ratio ) if (ref_matched[spread_key] > epsilon).any() and ( np.abs(np.mod(ref_matched[log2_key], 1)) > epsilon ).any(): # Pooled/paired reference only logging.debug("Weighting bins by coverage spread in reference") # NB: spread ~= SD, so variance ~= spread^2 fancy_wt = 1.0 - ref_matched[spread_key] ** 2 # Average w/ simple weights, giving this more emphasis x = 0.9 weights = x * fancy_wt + (1 - x) * simple_wt else: # Flat reference, only 1 weight estimate weights = simple_wt return cnarr.add_columns(weight=weights.clip(epsilon, 1.0))