Changelog¶

[0.4.10] (2026-03-06)¶

Changed¶

Reverted RSparsePro eigendecomposition optimization from v0.4.9 to preserve result consistency across all datasets

[0.4.9] (2026-03-06)¶

Changed¶

Optimized RSparsePro performance: replace repeated O(p³) matrix inversions with single eigendecomposition in adaptive_train loop, achieving 8-10x speedup for typical locus sizes

[0.4.8] (2026-03-03)¶

Changed¶

Optimized CLI startup speed by deferring heavy module imports (pandas, numpy, scipy, sklearn) and internal modules (credtools.credtools, credtools.locus, credtools.meta, credtools.qc) from module level to function level via lazy loading

[0.4.7] (2026-02-25)¶

Changed¶

Increased unit test coverage from 76% to 87% across 9 modules: credtools.py (59%→100%), credibleset.py (69%→97%), prepare.py (31%→95%), cli.py (34%→61%), plot.py (62%→75%), qc.py (85%→92%), validation.py (90%→97%), headers.py (90%→95%), RSparsePro.py (78%→100%)

Fixed¶

Applied black/isort formatting and resolved flake8 lint errors in test files

[0.4.6] (2026-02-25)¶

Added¶

Unit tests for susie_rss.py internal functions (coverage 16% → 82%): 107 tests covering credible sets, statistical gradients, SER, prior variance optimization, initialization, and integration
Unit tests for multisusie_rss.py internal functions (coverage 8% → 65%): 65 tests covering S/SER_RESULTS classes, recovery functions, compute_lbf, optimization, and integration

Changed¶

Upgraded typer>=0.9.0 and cojopy>=0.1.4 to fix click 8.2 incompatibility

Fixed¶

Added D102 and E226 to flake8 ignore list for pre-existing lint issues

[0.4.5] (2026-02-23)¶

Added¶

Outlier SNP detail tracking: identify_outliers() now returns a DataFrame with per-criterion flags (C1_ld_mismatch, C2_marginal, C3_dentist_s) instead of a plain SNPID list
outlier_snps.txt.gz output file saved at both per-locus and global levels, recording each outlier SNP with its triggered detection criteria
Updated QC documentation with detailed outlier detection criteria descriptions

Fixed¶

Use outer join in cochran_q() to include all SNPs across cohorts (previously only intersection)

[0.4.4] (2026-02-23)¶

Added¶

Comprehensive unit tests for all wrapper modules (93 new tests): ABF, ABF+COJO, FINEMAP, SuSiE, SuSiEx, MultiSuSiE, RSparsePro
--skip option to meta command to resume from previous runs

Fixed¶

Handle cohorts with no sumstats-LD overlap in QC metrics
Compute heterogeneity by population in meta_by_population mode
Format test files and fix D403 docstring lint errors

[0.4.3] (2026-02-22)¶

Fixed¶

Update uv.lock after version bump

[0.4.2] (2026-02-22)¶

Added¶

Comprehensive unit tests for QC module (tests/test_qc.py): 80 tests across 17 test classes covering get_eigen, estimate_s_rss, kriging_rss, compute_dentist_s, compare_maf, snp_missingness, ld_4th_moment, ld_decay, cochran_q, locus_qc, identify_outliers, remove_snps_from_locus, save_cleaned_locus, remove_outliers_and_rerun_qc, locus_qc_summary, qc_locus_cli, safe_qc_locus_cli, and loci_qc — achieving 82% coverage for credtools/qc.py

Changed¶

Rewrote docs/qc.md to match actual code: tabbed quick-start examples, Python API example using exampledata, complete CLI parameter reference, QC method descriptions (Kriging RSS, Dentist-S, MAF comparison), accurate output directory structure and column tables, mermaid pipeline diagram, collapsible troubleshooting FAQ, and note that heterogeneity metrics are now in credtools meta

[0.4.1] (2026-02-22)¶

Fixed¶

TestChunkDirect CI failures: Replaced dependency on gitignored exampledata/testout/ files with on-the-fly munge generation via munge_sumstats(), and replaced reference-data comparisons with self-contained assertions

[0.4.0] (2026-02-22)¶

Added¶

Comprehensive unit tests for meta module (tests/test_meta.py): 40 tests covering meta_sumstats, meta_lds, meta_all, meta_by_population, meta, meta_locus, meta_loci, and heterogeneity edge cases — achieving 100% coverage for credtools/meta.py

Changed¶

Rewrote docs/meta.md to match munge.md documentation style: tabbed examples, admonition blocks, collapsible FAQ, mermaid workflow diagram, typed column tables
Fixed outdated references in meta docs (prepared_loci_list.txt → chunked/loci_list.txt, incorrect output structure, wrong file extensions)
Added missing --calculate-lambda-s and --log-file CLI options to meta docs
Added Heterogeneity Analysis section documenting per-locus files and summary table columns

[0.3.10] (2026-02-21)¶

Added¶

Heterogeneity summary file (heterogeneity.txt.gz) with per-cohort aggregated metrics: ld_4th_moment_mean, ld_decay_rate, missing_rate, cochran_q_median, i_squared_median, n_het_snps
New heterogeneity_summary() function in meta module
Global heterogeneity summary across all loci in meta_loci() CLI flow
Chunk integration tests with exampledata

Changed¶

Moved heterogeneity computation from QC to meta module (computed before meta-analysis to capture pre-merge metrics)
meta_locus() now returns heterogeneity summary alongside results
save_heterogeneity() accepts optional summary parameter
pipeline() saves heterogeneity summary alongside detail files

Removed¶

docs/prepare.md and all references to credtools prepare

[0.3.9] (2026-02-15)¶

Added¶

Comprehensive unit tests for chunk command: 74 tests across 9 test classes covering credtools.preprocessing.chunk (100% coverage) and CLI chunk helper functions (_load_custom_chunks, _update_chunk_info_with_prepared, create_updated_chunk_info, parse_population_config_file)

[0.3.8] (2026-02-15)¶

Fixed¶

Replace hardcoded exampledata/testout/ paths in test_plot.py with pytest fixtures, fixing CI test failures when those files are not git-tracked

[0.3.7] (2026-02-13)¶

Fixed¶

Code formatting fixes (isort/black) and lint issue fixes (F821 undefined name, D104 missing docstring, D301 raw docstring)

[0.3.6] (2026-02-13)¶

Fixed¶

Fix test_run_fine_map_parallel: Renamed deprecated min_abs_corr parameter to purity and aligned mock return keys (causal_variants_records, cs_summary_records) with the actual _process_fine_map_task contract

[0.3.5] (2026-02-13)¶

Changed¶

Unified constants: Munging submodule now re-exports ColName, ColType, ColRange, ColAllowNA from main credtools.constants instead of maintaining duplicate definitions
Memory-efficient column types: Switched ColType to numpy types (np.int8, np.int32, np.float32, np.float64) for reduced memory footprint
Stricter NA policy: P, BETA, SE, OR, ORSE no longer allow NA values
Refactored sumstats munge functions: munge_chr, munge_bp, munge_allele, munge_pvalue, munge_beta, munge_se, munge_eaf, munge_maf now reuse validate_and_clean_column from the munging module
Conditional EAF/MAF processing: munge() in sumstats.py skips EAF/MAF steps when the EAF column is absent
Standardized RSID naming: Renamed ColName.RSID from "rsID" to "RSID" for consistency
Removed EAF from mandatory columns: ColName.mandatory_cols no longer requires EAF
Public transform API: transform_chr and transform_allele exposed as public functions in munging module

Added¶

Boundary behavior pinning tests for munge_bp, munge_chr, munge_pvalue, munge_se, munge_eaf

Removed¶

Duplicate ColName/ColType/ColRange/ColAllowNA classes from munging/constants.py
OR column from COMMON_COLNAMES alias mapping and fuzzy header matching
Dead commented-out PP_POLYFUN_* constants

[0.3.4] (2026-02-13)¶

Added¶

Comprehensive unit tests for munge command: 153 new tests covering the entire credtools/preprocessing/munging/ module
test_munging_constants.py — column definitions, ranges, and alias mapping (12 tests)
test_munging_validation.py — data validation and cleaning logic (30 tests)
test_munging_headers.py — separator detection, header mapping, fuzzy matching (27 tests)
test_munging_core.py — core munge pipeline, SNPID generation, allele transforms (38 tests)
test_munging_init.py — config I/O, load-and-munge integration (15 tests)
test_munge.py — CLI entry point, file handling, overwrite behavior (20 tests)
Shared test fixtures in tests/preprocessing/conftest.py
Coverage: constants 100%, core 98%, munge 95%, validation 85%, headers 79%

Changed¶

Improved munge documentation (docs/munge.md): rewritten with mkdocs-material admonitions, tabbed quick start, column auto-detection table, data cleaning rules, mermaid pipeline diagram, collapsible FAQ, and runnable example using bundled test data

[0.3.3] (2026-02-13)¶

Changed¶

Remove dead code and extract shared COJO/PIPs/CS-summary logic
Fix p-value numerical underflow for extreme Z-scores
Update CLI help text and dependencies
Enable internal purity filtering for SuSiE-based tools

[0.3.2] (2025-12-16)¶

Changed¶

Unified purity filtering: All fine-mapping tools now disable internal purity filtering (set to 0.0) and apply filtering uniformly at the credtools level
SuSiE, MultiSuSiE, and SuSiEx no longer perform tool-internal purity filtering
Purity filtering applied consistently after tool execution via filter_credset_by_purity()
Ensures identical filtering behavior across all 7 fine-mapping tools
User-facing --purity parameter behavior unchanged (backward compatible)

[0.3.1] (2025-12-16)¶

Fixed¶

QC threshold parameter consistency: Unified default threshold values between CLI and internal functions
Changed CLI defaults: logLR_threshold (1.0→2.0), z_threshold (1.0→2.0), r_threshold (0.1→0.8)
Fixed missing threshold parameters in remove_outliers_and_rerun_qc() causing inconsistent outlier detection
Cleaned data now correctly shows n_lambda_s_outlier = 0 when using same thresholds for detection and re-QC

[0.3.0] (2025-12-15)¶

Fixed¶

QC outlier removal output structure: Reorganized cleaned data output to improve clarity and downstream usability
BREAKING: All cleaned data now organized under {out_dir}/cleaned/ directory instead of {out_dir}/{locus_id}/cleaned/
Fixed missing files for loci without outliers: Previously, when no outliers were detected, cleaned data files were not saved, causing cleaned_loci_info.txt.gz paths to be invalid. Now all loci have cleaned data saved (identical to original if no outliers detected)
Added missing cleaned QC summaries:
- Per-locus cleaned QC summary: {out_dir}/cleaned/{locus_id}/qc.txt.gz
- Global cleaned QC summary: {out_dir}/cleaned/qc_cleaned.txt.gz

Changed¶

New output directory structure for cleaned data:

{out_dir}/
├── qc.txt.gz                              # Original QC summary
├── {locus_id}/                            # Original QC results
│   ├── qc.txt.gz
│   ├── expected_z.txt.gz
│   └── ...
└── cleaned/                               # NEW: All cleaned results
    ├── qc_cleaned.txt.gz                  # Global cleaned QC summary
    ├── outlier_removal_summary.txt.gz     # Global outlier removal summary
    ├── cleaned_loci_info.txt.gz           # Cleaned loci info for downstream
    └── {locus_id}/                        # Per-locus cleaned data
        ├── {prefix}.sumstats.gz
        ├── {prefix}.ld.npz
        ├── {prefix}.ldmap.gz
        ├── qc.txt.gz                      # Per-locus cleaned QC
        ├── expected_z.txt.gz
        └── outlier_removal_summary.txt.gz

Migration Guide¶

For users of `credtools qc --remove-outlier`¶

Old paths (no longer valid):

{out_dir}/qc.txt.gz
{out_dir}/outlier_removal_summary.txt.gz
{out_dir}/cleaned_loci_info.txt.gz
{out_dir}/{locus_id}/cleaned/{prefix}.sumstats.gz
{out_dir}/{locus_id}/cleaned/outlier_removal_summary.txt.gz

New paths:

{out_dir}/qc.txt.gz                                    # Original QC (unchanged)
{out_dir}/cleaned/qc_cleaned.txt.gz                    # NEW
{out_dir}/cleaned/outlier_removal_summary.txt.gz       # MOVED
{out_dir}/cleaned/cleaned_loci_info.txt.gz             # MOVED
{out_dir}/cleaned/{locus_id}/{prefix}.sumstats.gz      # MOVED
{out_dir}/cleaned/{locus_id}/qc.txt.gz                 # NEW
{out_dir}/cleaned/{locus_id}/outlier_removal_summary.txt.gz  # MOVED

For downstream fine-mapping: - Use {out_dir}/cleaned/cleaned_loci_info.txt.gz as input (path updated but still valid) - All paths in this file now correctly point to {out_dir}/cleaned/{locus_id}/

Benefits¶

Clearer organization: Original and cleaned data completely separated
Reliable paths: All cleaned data files guaranteed to exist, even when no outliers detected
Better QC tracking: Can compare qc.txt.gz vs cleaned/qc_cleaned.txt.gz to assess cleaning impact
Consistent structure: All cleaned outputs follow same directory pattern

[0.2.0] (2025-11-25)¶

BREAKING CHANGES¶

Renamed min_abs_corr parameter to purity across all CLI commands and Python API
CLI: --min-abs-corr flag replaced with --purity (short form: -p)
Python API: min_abs_corr= parameter renamed to purity= in all function signatures
Migration required: Update all scripts and workflows to use the new parameter name
Changed default purity threshold from 0.5/0.1 to 0.0 (no filtering by default)
Previous behavior: SuSiE family tools had min_abs_corr=0.1 (function) or 0.5 (CLI)
New behavior: purity=0.0 means no filtering unless explicitly set by user
Other tools (ABF, FINEMAP, RSparsePro) previously had no purity filtering
Migration impact: Users must explicitly set --purity to filter credible sets

Added¶

Unified purity calculation and filtering for all fine-mapping tools
All 7 tools now calculate and store purity values in CredibleSet objects
Purity definition: minimum absolute LD correlation between all variant pairs in a credible set
Tools now include: ABF, ABF+COJO, FINEMAP, RSparsePro (previously unsupported)
Multi-ancestry purity: Element-wise maximum LD across populations, then minimum
Centralized purity filtering logic
New filter_credset_by_purity() function in credibleset.py
Filtering applied in fine_map() and combine_creds() functions
Consistent behavior across all tools and analysis strategies
Enhanced combine_creds() function
Added min_purity parameter for filtering combined credible sets
Purity filtering applied after merging credible sets from multiple tools/loci

Changed¶

More intuitive parameter naming: "purity" more clearly describes the metric than "min_abs_corr"
Opt-in filtering model: Users choose when to apply purity filtering (default: no filtering)
Transparent purity values: All tools now expose purity calculations for user inspection

Migration Guide¶

For CLI Users¶

# Old command (no longer works)
credtools finemap --min-abs-corr 0.5 [other options]

# New command
credtools finemap --purity 0.5 [other options]

# Or use short form
credtools finemap -p 0.5 [other options]

For Python API Users¶

# Old code (no longer works)
from credtools import fine_map
result = fine_map(locus_set, tool="susie", min_abs_corr=0.5)

# New code
from credtools import fine_map
result = fine_map(locus_set, tool="susie", purity=0.5)

Recommended Actions¶

Update all scripts: Replace --min-abs-corr with --purity and min_abs_corr= with purity=
Review filtering needs: Since default changed to 0.0 (no filtering), explicitly set --purity if needed
For old SuSiE behavior: Use --purity 0.1 to match previous SuSiE function default
Check purity values: All tools now store purity - inspect credset.purity for quality assessment

Benefits¶

Consistency: Same purity filtering available for all 7 fine-mapping tools
Flexibility: Users control filtering threshold per analysis
Transparency: Purity values accessible for all credible sets
Better quality control: Unified metric for assessing credible set reliability

[0.1.0] (2025-11-25)¶

Added¶

Adaptive max_causal for multi-input tools: Extended adaptive max_causal parameter tuning to multi-input fine-mapping tools (multisusie and susiex)
Added _adaptive_fine_map_multi() function for LocusSet-level adaptive logic
Same algorithm as single-input tools: increase max_causal by 5 (up to 20) when saturated, decrease by 1 (down to 1) on failure
Multi-input tools now support --adaptive-max-causal flag for automatic parameter optimization
Unified adaptive strategy across all supported fine-mapping tools

Changed¶

SuSiEx mult_step default: Changed default value from True to False to avoid conflicts with credtools' adaptive logic
Users can still enable SuSiEx's internal multi-step refinement by explicitly setting --mult-step
Updated documentation to clarify interaction between --mult-step and --adaptive-max-causal

Enhanced¶

Documentation improvements:
Updated adaptive_max_causal parameter documentation to include multi-input tool support
Enhanced CLI help text for --mult-step option with usage guidance
Added detailed docstrings explaining adaptive logic for both single-locus and LocusSet processing

Technical Details¶

Single-input tools (finemap, susie, rsparsepro): Adaptive logic applies per-locus
Multi-input tools (multisusie, susiex): Adaptive logic applies to entire LocusSet
Backward compatible: adaptive_max_causal defaults to False, existing workflows unchanged

[0.0.41] (2025-11-22)¶

Fixed¶

Multi-cohort CS purity calculation: Fix ValueError when calculating credible set purity across cohorts with non-overlapping SNP sets. The function now takes the union of all CS SNPs across cohorts and expands each cohort's LD matrix to the union size (following MultiSuSiE's approach), with missing SNPs represented as uncorrelated (LD=0, diagonal=1). This resolves broadcasting errors like "operands could not be broadcast together with shapes (96,96) (95,95)".

[0.0.40] (2025-10-30)¶

Added¶

add cred and PIP of each cohort in multi cohort fine-mapping
output zero credible sets when n_sig 5e-8 = 0

Fixed¶

fix multisusie doesn't filter purity issue

[0.0.39] (2025-09-22)¶

Added¶

Parallel FINEMAP execution: credtools finemap accepts --processes/-np to fan out loci across worker processes, keeping sequential behaviour available with the default of 1.

Changed¶

Per-locus task coordination: Fine-mapping runs now stream through a worker-safe task wrapper, preserving rich progress reporting, capturing tracebacks, and writing locus outputs from whichever worker completes first.
CLI artefacts: Run summaries and combined credible set exports remain deterministic under parallel execution, with parameters.json recording the requested worker count.

Tests¶

Added regression coverage for the CLI multiprocessing path by mocking the worker pool to assert identical outputs for 1 vs N processes.

[0.0.38] (2025-09-21)¶

Added¶

FINEMAP timeout control: credtools finemap now accepts --timeout-minutes (default 30) to cap per-locus runs, with timeout failures reported in run_summary.log.

Changed¶

External tool execution: FINEMAP wrapper and tool manager enforce the timeout and append [timeout] markers to logs when the limit is exceeded, ensuring stalled runs surface immediately.

[0.0.37] (2025-09-20)¶

Fixed¶

LD matrix ingestion: Replace NaN entries with zeros when loading from lower-triangle text or .npz files, and add regression tests to guard behavior.
QC pipeline robustness: Locus-level exceptions no longer abort credtools qc; failures are logged per locus, success/failure counts are summarized to qc_run_summary.log, and CLI feedback reflects mixed outcomes.
Error messaging: Intersecting sumstats/LD now reports the offending locus when common variants are missing to simplify debugging.

[0.0.36] (2025-09-18)¶

Changed¶

CLI preprocessing flows: credtools munge and credtools chunk now accept comma- or newline-separated input paths in addition to TSV config files, with friendlier validation and logging when LD references are omitted.
Preprocessing outputs: Munge results retain the sample-size N column in the standard export schema and downstream validators require it.
Optional plotting dependency: Centralized the upsetplot import so the plotting module gatekeeps the dependency once and falls back gracefully when the package is absent.

[0.0.35] (2025-09-17)¶

Added¶

Comprehensive Plotting Module: New visualization capabilities for QC results
Added credtools.plot module with publication-quality plotting functions
Summary QC Plots (2x2 layout from qc.txt.gz):
- Lambda-s distribution boxplot by cohort
- MAF correlation barplot between summary statistics and LD reference
- Lambda-s outliers count barplot by cohort
- Dentist-s outliers count barplot by cohort
Locus-specific Plots (2x2 layout from individual QC files):
- Locus p-value plot with credible set annotations (from expected_z.txt.gz)
- Observed vs expected z-scores QQ plot (from expected_z.txt.gz)
- LD decay curve plots by cohort (from ld_decay.txt.gz)
- LD 4th moment boxplots by cohort (from ld_4th_moment.txt.gz)
- SNP missingness upset plot showing overlap patterns (from snp_missingness.txt.gz)
CLI Command:
- credtools plot: Unified plotting command with auto-detection of plot type
- Supports all plot types: summary (2x2), locus (2x2), and individual plots
- Smart auto-detection based on input path (directory for locus, qc.txt.gz for summary)
Features:
- Population-aware color schemes for consistent visualization
- Support for PNG, PDF, SVG output formats
- Customizable figure sizes and DPI settings
- Graceful handling of missing files and optional dependencies
- Professional styling with seaborn integration

Dependencies¶

Added seaborn>=0.11.0 for enhanced statistical plotting
Added upsetplot>=0.6.0 for intersection visualization (optional for SNP missingness plots)

Enhanced¶

QC workflow now generates comprehensive visualizations alongside numerical results
All plotting functions integrate seamlessly with existing QC output file formats
Error handling with informative messages for missing dependencies or data files

[0.0.34] (2025-09-14)¶

Fixed¶

CI Test Suite: Fixed failing CI tests and errors
Fixed TypeError in test fixtures by adding required locus_start and locus_end parameters to Locus constructor calls
Removed unused toml import that was causing ModuleNotFoundError in integration tests
Fixed linting issues: converted f-strings without placeholders to regular strings
Added missing docstrings for mock classes in QC module
All 70 tests now pass successfully

Improved¶

Enhanced test coverage and reliability of CI pipeline
Improved code quality with resolved linting and documentation issues

[0.0.33] (2025-09-12)¶

Added¶

QC Summary Statistics: Enhanced quality control with comprehensive summary reports
Added locus-level QC summary files (qc.txt.gz) in each locus directory
Added global QC summary file (qc.txt.gz) in output root directory
Summary includes key QC metrics: SNP counts, significance thresholds, MAF correlations, lambda-s values, and outlier counts
Configurable thresholds for flip detection, lambda-s outliers, and Dentist-S outliers
New locus_qc_summary() function to generate summary statistics from detailed QC results
Enhanced QC metrics:
- n_1e-5, n_5e-8: Count of SNPs below significance thresholds
- maf_corr: Correlation between summary statistics and LD reference MAF
- n_flip: Count of potential allele flips (logLR > 2 AND |z| > 2)
- n_lambda_s_outlier: Count of lambda-s outliers (|z_std_diff| > 3)
- n_dentist_s_outlier: Count of Dentist-S outliers (-log10p ≥ 4 AND r² ≥ 0.6)

Changed¶

Modified locus_qc() function to accept threshold parameters for outlier detection
Updated qc_locus_cli() to return both locus ID and summary statistics
Enhanced loci_qc() function to aggregate and save global QC summary across all loci
Improved QC workflow output structure with hierarchical summary files

[0.0.32] (2025-09-12)¶

Changed¶

BREAKING: Removed strategy parameter from fine-mapping interface
Fine-mapping strategy is now automatically determined based on tool type and data structure
Multi-input tools (susiex, multisusie) automatically process all loci together
Single-input tools automatically combine results when multiple loci are provided
Added deprecation warning for backward compatibility
Enhanced CLI with enum validation for combination methods
Added CombineCred enum for credible set combination methods (union, intersection, cluster)
Added CombinePIP enum for PIP combination methods (max, min, mean, meta)
Improved input validation and auto-completion support

Removed¶

Web visualization feature moved to v2 (will be available in future release)
Removed credtools web command documentation
Removed web-related installation instructions
Removed web tutorial files and examples
Updated all workflow examples to reference output files instead

Improved¶

Simplified user interface with automatic strategy selection
Better CLI help with enum option display
Updated documentation to reflect streamlined workflow

[0.0.31] (2025-09-11)¶

Fixed¶

CI error

[0.0.30] (2025-09-11)¶

Added¶

ABF+COJO
adaptive causal

[0.0.28] (2025-06-13)¶

Added¶

add api docs

[0.0.27] (2025-06-12)¶

Added¶

add set_L_by_cojo to cli:pipeline

[0.0.26] (2025-06-02)¶

Added¶

add web visualization

[0.0.25] (2025-06-02)¶

Added¶

add tutorial

[0.0.23] (2025-02-01)¶

Fixed¶

fix finemap cred bug

[0.0.21] (2025-01-20)¶

Fixed¶

fix no install error for carma

[0.0.20] (2025-01-20)¶

Fixed¶

fix zero maf in finemap

[0.0.19] (2025-01-20)¶

Added¶

qc support for multiprocessing

[0.0.18] (2025-01-19)¶

Fixed¶

fix the bug of no credible set

[0.0.17] (2025-01-18)¶

Added¶

support for multiprocessing
add progress bar

[0.0.16] (2025-01-18)¶

Added¶

support for sumstats.gz and ldmap.gz

[0.0.15] (2024-12-17)¶

Added¶

cli args

[0.0.14] (2024-12-16)¶

Added¶

cli

[0.0.13] (2024-12-16)¶

Added¶

pipeline

[0.0.12] (2024-12-15)¶

Added¶

ensemble fine-mapping

[0.0.11] (2024-12-15)¶

Added¶

multisusie

[0.0.10] (2024-12-15)¶

Added¶

susiex
Rsparseld
CARMA

[0.0.9] (2024-10-21)¶

Added¶

abf
susie
finemap

[0.0.8] (2024-10-10)¶

Added¶

load ld matrix and ld map
munge sumstat
example data

[0.0.7] (2024-10-09)¶

Added¶

test for ldmatrix

[0.0.6] (2024-10-09)¶

Added¶

functions for load LD
test for ColName

[0.0.5] (2024-10-08)¶

First release on PyPI.

Changelog¶

[0.4.10] (2026-03-06)¶

Changed¶

[0.4.9] (2026-03-06)¶

Changed¶

[0.4.8] (2026-03-03)¶

Changed¶

[0.4.7] (2026-02-25)¶

Changed¶

Fixed¶

[0.4.6] (2026-02-25)¶

Added¶

Changed¶

Fixed¶

[0.4.5] (2026-02-23)¶

Added¶

Fixed¶

[0.4.4] (2026-02-23)¶

Added¶

Fixed¶

[0.4.3] (2026-02-22)¶

Fixed¶

[0.4.2] (2026-02-22)¶

Added¶

Changed¶

[0.4.1] (2026-02-22)¶

Fixed¶

[0.4.0] (2026-02-22)¶

Added¶

Changed¶

[0.3.10] (2026-02-21)¶

Added¶

Changed¶

Removed¶

[0.3.9] (2026-02-15)¶

Added¶

[0.3.8] (2026-02-15)¶

Fixed¶

[0.3.7] (2026-02-13)¶

Fixed¶

[0.3.6] (2026-02-13)¶

Fixed¶

[0.3.5] (2026-02-13)¶

Changed¶

Added¶

Removed¶

[0.3.4] (2026-02-13)¶

Added¶

Changed¶

[0.3.3] (2026-02-13)¶

Changed¶

[0.3.2] (2025-12-16)¶

Changed¶

[0.3.1] (2025-12-16)¶

Fixed¶

[0.3.0] (2025-12-15)¶

Fixed¶

Changed¶

Migration Guide¶

For users of credtools qc --remove-outlier¶

Benefits¶

[0.2.0] (2025-11-25)¶

BREAKING CHANGES¶

Added¶

Changed¶

Migration Guide¶

For CLI Users¶

For Python API Users¶

Recommended Actions¶

Benefits¶

[0.1.0] (2025-11-25)¶

Added¶

Changed¶

Enhanced¶

Technical Details¶

[0.0.41] (2025-11-22)¶

Fixed¶

[0.0.40] (2025-10-30)¶

Added¶

Fixed¶

For users of `credtools qc --remove-outlier`¶