Merge branch 'deps_upgrade' into 'dev'

Deps upgrade: bedtools==2.31; porechop==artic; pandas>=2.0 support

See merge request research/pomoxis!167

Author: ftostevin-ont

CHANGELOG.md

@@ -4,15 +4,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-## Unreleased
-### Added
-- Option `-C` for `mini_align` to copy fastx comments into bam tags
-
-## [v0.3.13] - 2023-06-12
+## [v0.3.13] - 2023-06-23
 ### Changed
 - `subsample_bam`: `--force_low_coverage` saves contigs with coverage below the target 
 - `subsample_bam`: `--force_non_primary` saves multimapping for the subsampled reads
 - `coverage_from_bam`: `--primary_only` considers only primary reads when computing the depth
+- `bedtools`: upgraded to v2.31
+- `porechop`: switched to using Artic version 
+### Added
+- Option `-C` for `mini_align` to copy fastx comments into bam tags
+### Fixed
+- Minor compatibility fixes to support `pandas>=2.0` 
+
 ## [v0.3.12] - 2023-02-09
 ### Changed
 - `subsample_bam`: `--quality` filtering now uses mean error probability, not mean of quality scores as previously.

Makefile

@@ -121,8 +121,8 @@ $(BINCACHEDIR)/seqkit: | $(BINCACHEDIR) $(BINBUILDDIR)
 	cp ${BINBUILDDIR}/seqkit $@	
 
 
-#BEDTOOLSVER=2.29.0 Changed to 2.30.0 on 6 December 2021
-BEDTOOLSVER=2.30.0
+#BEDTOOLSVER=2.31.0 on 7th June 2023
+BEDTOOLSVER=2.31.0
 $(BINCACHEDIR)/bedtools: | $(BINCACHEDIR) $(BINBUILDDIR)
 	@echo Making $(@F)
 	if [ ! -e ${BINBUILDDIR}/bedtools-2.29.0.tar.gz	]; then \

pomoxis/assess_homopolymers.py

@@ -127,7 +127,7 @@ def plot_relative_lengths(data, fname):
     fig, axes = plt.subplots(
         ncols=len(cols), nrows=len(rows), sharex=True,
         figsize=(4 * len(cols), 2 * len(rows)), squeeze=False)
-    for rl, rl_df in data.groupby(['ref_len']):
+    for rl, rl_df in data.groupby('ref_len'):
         i = rows.index(rl)
         for qb, qb_df in rl_df.groupby('q_base'):
             j = cols.index(qb)
@@ -168,7 +168,7 @@ def get_heatmap(data, data_comp):
     df_comp[np.isnan(df_comp)] = 0
     df_comp = df_comp.reindex(column_reindex, axis=1)
 
-    df_all = df_comp.sum(level='ref_len', axis=1)
+    df_all = df_comp.groupby(level='ref_len', axis=1).sum()
     df_all.columns = pd.MultiIndex.from_product((['ATGC',], df_all.columns), 
         names=['q_base', 'ref_len'])
 

pomoxis/catalogue_errors.py

@@ -703,7 +703,7 @@ def plot_summary(df, outdir, prefix, ref_len):
     fig.subplots_adjust(left=0.3)
     y_pos = np.arange(len(df) + 1)
     no_error_score = -10 * np.log10(1/ref_len)
-    ax.barh(y_pos, df['remaining_err_rate_q'].append(pd.Series(no_error_score)), align='center', color='green', ecolor='black')
+    ax.barh(y_pos, np.concatenate([df['remaining_err_rate_q'].to_numpy(), [no_error_score]]), align='center', color='green', ecolor='black')
     ax.set_xlabel('Q(Accuracy)')
     ax.set_ylabel('Error Class')
     ax.set_ylim((y_pos[0]-0.5, y_pos[-1]+0.5))

requirements.txt

@@ -7,4 +7,4 @@ ncls
 numpy
 pandas
 pysam >= 0.15.4
-git+https://github.com/rrwick/Porechop
+porechop @ git+https://github.com/artic-network/porechop