Update Logs

$ 3.5.1 2024 - fixed errors due to liftover version change

$ 3.5.0 20241029 - added plot_gwheatmap() - added genename annotation in compare_effect() - fixed annotation error in compare_effect() - fixed is_q error in compare_effect() - fixed error for calculating arm length for annotation. - removed some outdated code

3.4.49 2024

• fixed the suffix for yaml in to_format()
• implemented .flip_snpid and .strip_snpid
• implemented support for chm13 13 in .liftover
• updated reference url list (added chain files)
• implemented chrom_pat and snpid_pat filters for gl.Sumstats()
• fixed bug in .plot_miami2
• updated reference url

3.4.48 20240822

• fixed bug for is_q_mc in compare_effect()
• added anno_height for plot_mqq()
• added xtight for plot_mqq()
• fixed bug for xpad in plot_mqq()
• supported baseline model for ldsc outputs

3.4.47 20240703

• updated stacked regional plot
• supported user-provided plink path
• fixed a bug in column order for to_format()

3.4.46 20240624

• added version requirements for numpy (<2) and matplotlib (<3.9)

3.4.45 20240509

• changed python version requirements to >=3.9, <3.11
• updated the version of pysam to v0.22.1
• fixed bug for status code when flipping statistics

3.4.44 20240424

• fixed bug when POS > sequence length for check_ref()
• vectorized normalize_allele()

3.4.43 20240403

• Added cache to speed up strand inference (credit to @sup3rgiu Mr. Andrea)
• fixed error for .plot_mqq(m="qq")
• Added h5py==3.10.0 to dependencies.

3.4.42 20240328

• fast implementation of check_ref and to_format. (credit to @sup3rgiu Mr. Andrea)
• added highlight and pinpoint for plot_trumpet()
• fixed typo (credit to @sup3rgiu Mr. Andrea)
• changed **args to **kwargs (credit to @sup3rgiu Mr. Andrea)
• implemented munge-like filters munge=True for ldsc in gwaslab

3.4.41 20240219

• fixed error in region_ref_second

3.4.40 20240215

• fixed color issue for regional plots. The color assigned to each variant is actually the color for the lower LD r2 category. For example, variants with LD>0.8 will be colored with the color for 0.8>LD>0.6.
• integrated LDSC (partitioned h2/h2-cts)
• added wc_correction for get_lead()
• updated log

3.4.39 20240207

• integrated LDSC h2/rg
• updated LICENSE from MIT to GPL-3.0 license
• added check_novel_set / check_cis
• added filter_snp/palindromic/indel and filter_hapmap3
• updated log system
• fixed bug when gene name is empty in GTF in regional plot
• fixed error in harmonization when there is no palindromic SNPS or indels to check.
• fixed bug when saving plot as pdf with matplotlib>3.6
• fixed typos

3.4.38 20240203

• fixed seaborn version
• replaced array_split
• fixed warnings due to pandas upgrade
• renamed get_flanking to filter_flanking
• fixed error in cbar for regional plot
• restructured log system
• restructured sanity checking
• restructured stats flipping
• updated tutorial and examples
• fixed error in log

3.4.37 20240129

• added data consistency check
• added "s","r","n" mode for remove_dup()
• updated fix_id
• added sample data
• added datatype check for rsID and SNPID
• added memory usage check
• added extreme P value check

3.4.36 20240126

• removed statsmodels
• updated functions in basic_check()
• added a test example for basic_check()
• fixed fontsize and font_family errors in plot_mqq() for ax4
• added new options for plot_mqq(mode="r"); cbar_fontsize, cbar_font_family, cbar_title
• added new options for plot_mqq(mode="r"); track_n, track_n_offset, track_fontsize_ratio, track_exon_ratio, track_text_offset, track_font_family

3.4.35 20240123

• fixed version number

3.4.34 20240123

• get_lead will use MLOG10P first instead of P (scaled is deprecated in get_lead)
• added datatype check for fill_data
• added datatype verification
• updated sanity check default values (BETA, OR, HR)

3.4.33 20240122

• fix_allele() now prints out allele information
• updated sanity check default values (added tolerance for floats)
• fixed error in loading pickle created by older versions
• fixed bug in flipping OR and HR
• added version information to the start line of each checking function
• updated reference VCF
• updated clump() / to_finemapping() / run_susie_rss() (beta)
• updated Manhattan-like plotting system
• updated datatype for certain statistics
• updated file naming system (beta)
• added stacked mqq plot (beta)
• plot_miami2() can iteratively call plot_mqq to create miami plot, which supports more functions (beta)

3.4.32 20231118

• fixed bug in sanity_check : N = N_CASE + N_CONTROL
• fixed the logic chain for normalization
• added extra log in compare_effect
• fixed a few typos (credit to @gmauro Mr. Gianmauro Cuccuru)
• added clump() for beta testing
• added to_finemapping() for beta testing
• added run_susie_rss() for beta testing

3.4.31 20231115

• updated version requirements

3.4.30 20231103

• updated random seed range (0,2^32-1).
• fixed error for suggestive_sig_line in miami plot.
• loosened version requirements for python, pandas and matplotlib.

3.4.29 20231003

• set scatter_kwargs={"rasterized":True} as default for miami plot

3.4.28 20231003

• fixed a bug in compare_effect

3.4.27 20231002

• update reference book
• added extra parameters for tabix_index()
• update random variants

3.4.26 20230911

• added infer_af()
• updated reference datasets

3.4.24 20230821

• supported highlight and pinpoint for multiple sets of loci and variants in .plot_mqq()
• added overwrite option for gl.download_ref()

3.4.24 20230821

• update references

3.4.23 20230817

• fixed bug in get_lead(). In some rare cases, it was not counted when the last variant is a new lead variant.

3.4.22 20230803

• added plot_power() and plot_power_x()
• added support for multiple EFO IDs
• updated built-in formatbook
• fixed regex error in read_ldsc for numbers in the format of 1e-01
• fixed typos

3.4.21 2023/07/18

• fix error in gc calculation for mqq plot when expected_min_mlog10p is not 0 and stratified=True.

3.4.20 2023/07/15

• reimplement and unified the module for saving figures
• added trumpet plot document page
• added CHR range check in .fix_chr() and .plot_mqq() (remove variants with CHR<=0) (#42)
• fixed error in column headers for plot_mqq "b" mode (#40)
• fixed many typos.. (#41)
• fixed error in plot_mqq logging (warning for genome build)
• fixed error in annotation for miami plot

3.4.19 2023/06/29

• added annotation using custom column for miami plot
• updated alogorithm for extracting lead variants (using scaled P)

3.4.18 2023/06/28

• fixed bugs for miami plot

3.4.17 2023/06/27

• added expected_min_mlog10p for qq mode in plot_mqq()
• added trumpet plot
• added support for gwaslab Sumstats object for gl.compare_effect()
• added saving options for gl.compare_effect()
• fixed bug for is_q=False in gl.compare_effect().

3.4.16 2023/06/22

• fixed bug for qq mode in plot_mqq()

3.4.15 2023/06/20

• LDSC-rg genetic correlation heatmap
• Allele frequency correlation plot
• Miami plot using gl.Sumstats Object pickle files
• Auto-check for VCF chromosome prefix (chr1 or 1)
• fill_data() is now implemented iteratively
• Downloaded files auto detection.

3.4.14 2023/06/09

• fixed bug

3.4.13 2023/06/09

• added two-reference-variant mode for regional plot
• auto-check for genome build version
• added additional_line and additional_line_color for plot_mqq
• fixed bugs for sig_level_lead

3.4.12 2023/05/30

• added highlight_anno_args in plot_mqq()
• added GWAS-SSF style metadata yaml output support
• added region_ref for regional plots
• updated default formatbook

3.4.11 - 2023/04/28

• fixed anno=None in miami plot

3.4.10 - 2023/04/27

• fixed variant_id sep for ssf
• update tutorials

3.4.9 - 2023/04/25

• update variant matching criteria for regional plot
• fixed default values for mqqplot
• added qq_scatter_args
• added connection timeout, status code check and md5sum verification for download_ref

3.4.8 - 2023/04/21

• fixed regional plot lead variant line error
• added tutorial for v3.4
• fixed anno_alias priority

3.4.7 - 2023/04/20

• fixed font_family for annotation
• get_lead can now use mlo10p to extract lead variants
• added sig_level_lead in plot_mqq

3.4.6 - 2023/04/18

• updated yticklabel fontsize
• updated drop_chr_start

3.4.5 - 2023/04/06

• added rr_lim : input a tuple like (0,100) or "max"

3.4.4 - 2023/04/04

• fixed error in mqqplot (chr > 26)

3.4.3 - 2023/03/29

• added jagged y-axis
• added cut_log

3.4.2 - 2023/03/28

• fixed y tick labels
• added font_family
• added ylabels

3.4.1 - 2023/03/28

• added sc_linewidth
• fixed use_rank
• fixed ystep

3.4.0

• restructured plot functions
• added dtype conversion for input pd.DataFrame
• reimplemented gtfparse and revised requirements
• update reference datasets
• fixed bugs when no variants were selected for mqqplot
• fixed bugs in gl.check_downloaded_ref()
• added suggestive significance line

3.3.24 - 2023/02/01

• update bugs in init.py

3.3.23 - 2023/01/31

• update annotation arrow style anno_style for .plot_mqq
• fixed effect size comparison bugs (added sorting)
• added allele check for effect size comparison
• update large number selection algorithm
• fixed dtype errors in fix_chr and fix_pos
• added perSNPh2 .get_per_snp_r2() and F statistics
• implemented save in Miami plots
• fixed annotation error in Miami plots

3.3.22 - 2023/01/27

• added checks for duplicates and NAs in compare_effect()

3.3.21 - 2023/01/27

• implement package info gl.show_version()
• fixed rsid_to_chrpos()

3.3.20

• updated get_density() to calculate the signal density for sumstats
• implemented winner's curse correction for effect size comparison

3.3.19

• updated fontsize options for plot_mqq(). Added anno_fontsize, title_fontsize.
• updated default values and optimized methods for remove_dup(), fix_chr(), basic_check(),check_sanity().
• added dump_pickle() load_pickle() to save half-finished sumstats object.

3.3.18

• updated config and downloading system

3.3.17

• added xtcik_chr_dict
• fixed bugs in miami plot

3.3.16

• added hg38 recombination rate file
• fixed bugs in get_novel

3.3.15

• fixed bugs for reading gtf files

3.3.14

• updated requirements for dependencies
  • pandas>=1.3,<1.5
  • pyensembl==2.2.3
• support customized gtf/vcf/recombination_rate files
• fixed bugs for regional plot
• calculate_gc()
• fill MAF : .fill_data(to_fill=["MAF"])

3.3.13

• specified python engine for query

3.3.12

• fixed bugs for matplotlib v3.6.x
• added method chain for filter_xxx functions
• updated requirements for dependencies
  • pySAM>=0.18.1,<0.20
  • matplotlib>=3.5
  • pyensembl>=2.2.3

3.3.11

• updated requirements for dependencies

3.3.10

• updated bugs for mqqplot

v3.3.9

• updated download system

v3.3.8

• included recombination data

v3.3.7

• updated packaging methods. Now when installing gwaslab, pip will install all dependencies as well.

v3.3.6 - 2022/11/05

• added download function:
  • now you can download reference files from predefined list via gwaslab
  • gl.check_available_ref() : list available reference files
  • gl.check_downloaded_ref() : list downloaded reference files
  • gl.download_ref(name) : download reference files
  • gl.remove_file(name) : remove the local reference files
  • gl.get_path(name) : get the local path for the reference data name
• implemented parsing gwas-vcf (fmt="vcf")
• implemented Sumstats.filter_value(expr)
• fixed bugs for check_allele
• optimized functions for sorting columns
• removed outdated codes in Sumstats

v3.3.5 - 2022/11/02

• added filter_value
• integrate gwascatalog to get_novel
• optimized remove_dup
• fixed bugs

v3.3.4 - 2022/10/31

• added gwascatalog_trait()
• optimized check_sanity()
• optimized the logic for removing duplicated and multiallelic variants
• added update_formatbook()
• added functions to read vcf.gz gl.Sumstats("myvcf.vcf.gz",fmt="vcf")
• gwaslab is now able to read chromosome-separated files
• fixed bugs

v3.3.3

• added Miami plot
• added Brisbane plot
• updated tutorials
• mqq plot annotation: new customization options

v3.3.2 - 2022/10/21

• added forcefixid for fix_id()
• fixed bugs for plotting gene tracks

v3.3.1

• extract novel loci given a list of known lead variants
• fixed bugs in fill_data()
• fixed path for hapmap3 snps for infer_build()

v3.3.0 - 2022/10/18

• added forest plot
• fixed options for mqqplot
• supported vcf

v3.2.0

• incorporated pyensembl and scikit-allel.
• get_lead() : support automatic gene name annotation (using pyensembl)
• to_format():
  • support common sumstats formats
  • support 1-based bed-like formats for VEP
  • support 0-based bed-like formats
• manhattan plot:
  • optimized plotting logic
  • annotate gene names
  • added regional plot feature using a user-provided reference panel
• comparison effect plot:
  • fix using OR

v3.1.0

• implemented formatbook: easily import sumstats and output sumstats in certain formats (support for commonly used formats including ldsc, plink, plink2, gwas-ssf, saige, regenie, fastgwa, metal, mrmega, pgscatalog, pgscatalog_hm, gwascatalog, gwascatalog_hm and gwaslab)
• added .filter_region_in/out using bed files (or in-built regions like high-ld or hla)
• implemented .summay() methods.
• optimized rsID annotation pipeline. Support annotation using curated chr:pos:ref:alt - rsID tsv for quick annotation.
• changed some datatypes and optimized memory usage.
• replaced pyVCF with pySAM