help menu tweaks

Author: Mike Lee

CHANGELOG.md

@@ -13,7 +13,7 @@
 
 -->
 
-## v1.14.0 (NOT RELEASED YET)
+## v1.14.0 (07-Apr-2026)
 
 ### Added
 - to `bit-cov-analyzer`
@@ -33,6 +33,7 @@
   - default window size change from 50 to 100, and default step size changed from 10 to 20
   - drastic improvements to efficiency when working with large genomes (e.g., 3GB)
   - histogram of coverages no longer plotted by default, only done now when adding the `--write-window-stats` flag
+  - no longer produces window-coverage-overview.txt as all of that info is captured within window-coverage-overview.tsv
 - `bit-get-mapped-reads-pid`
   - minor improvements to efficiency
 - `bit-get-cov-stats`

bit/cli/cov_analyzer.py

@@ -6,17 +6,17 @@
 def main():
 
     desc = """
-        This script identifies and pulls out regions of zero-coverage and relatively higher and lower coverage when
-        given a reference fasta and a bam file. It generates a bed file of the specified window-
-        and step-size, utilizes mosdepth to get the coverage of those windows, then generates stats
-        for those windows and pulls out regions with zero coverage and coverage above and below specified thresholds.
-        It outputs a table of coverage stats for all windows (if requested), a table of merged adjacent windows
-        ("regions"), and identified regions in fasta format. By default it looks for coverage variations
-        from the global mean coverage, but you can tell it to use per-contig coverages by adding the
-        `--per-contig` flag. Additionally, it is recommended to exclude contigs holding mitochondrial
-        genomes or chloroplasts due to their generally very relative high coverage, unless you
-        specifically want to investigate them too (and if so, you should probably use `--per-contig` mode). For
-        version info, run `bit-version`.
+        This program analyzes coverage patterns given a reference fasta and a bam file as inputs.
+
+        It generates a bed file of the specified window- and step-size, utilizes mosdepth
+        to get the coverage of those windows, then generates stats for those windows and
+        pulls out regions with zero coverage and regions with coverage above and below specified thresholds.
+        Primary outputs include tables and fastas of all merged adjacent windows ("regions"). By default, it
+        looks for coverage variations from the global mean coverage, but you can tell it to use per-contig
+        coverages by adding the `--per-contig` flag. Additionally, it is recommended to exclude contigs
+        holding mitochondrial genomes or chloroplasts due to their relatively high coverage (if you
+        specifically want to investigate them too, you should probably use `--per-contig` mode).
+        For version info, run `bit-version`.
         """
 
     parser = argparse.ArgumentParser(
@@ -95,7 +95,7 @@ def main():
         "-w",
         "--window-size",
         metavar="<INT>",
-        help='Sliding window size (default: 100)',
+        help='Window size (default: 100)',
         type=int,
         default=100,
     )
@@ -111,21 +111,21 @@ def main():
         "-g",
         "--allowed-gap",
         metavar="<INT>",
-        help='Number of bases allowed between qualifying windows (those with coverages above/below the specified thresholds) to still merge them into one contiguous region (default: 1000)',
+        help='Number of bases allowed between qualifying windows to still merge them into one contiguous region (default: 1000)',
         type=int,
         default=1000,
     )
     optional.add_argument(
         "-B",
         "--buffer",
         metavar="<INT>",
-        help='Add this length to each side of a high/low region of interest when pulled out as fasta (default: 100)',
+        help='Add this length to each side of a region of interest when pulled out as fasta (default: 100)',
         type=int,
         default=100,
     )
     optional.add_argument(
         "--write-window-stats",
-        help='Add this flag to also write out individual window stats and histogram of coverages (saves spacetime by not doing so)',
+        help='Add this flag to also write out individual window stats and a histogram of coverages (saves spacetime by not doing so)',
         action="store_true",
     )
 

bit/modules/cov_analyzer.py

@@ -563,7 +563,6 @@ def generate_outputs(reference_fasta, high_merged_regions, low_merged_regions,
 
 
 def write_window_cov_stats(cov_stats, output_dir, contig_lengths, log_file):
-    out_txt = f"{output_dir}/window-coverage-overview.txt"
     out_tsv = f"{output_dir}/window-coverage-overview.tsv"
     percentiles = [0.01, 0.1, 1, 5, 10, 25, 50, 75, 90, 95, 99, 99.9, 99.99]
 
@@ -598,58 +597,12 @@ def write_window_cov_stats(cov_stats, output_dir, contig_lengths, log_file):
             **{f"p{p}": group["cov"].quantile(p/100) for p in percentiles}
         })
 
-    # writing assembly level
-    with open(out_txt, "w") as out:
-        out.write("====================================\n")
-        out.write("=== GLOBAL WINDOW COVERAGE STATS ===\n")
-        out.write("====================================\n")
-        out.write(f"Assembly size: {assembly_length:,} bp\n")
-        out.write(f"Number of windows: {len(cov_stats.df)}\n")
-        out.write(f"Mean:   {cov_stats.global_mean:.2f}\n")
-        out.write(f"Median: {cov_stats.global_median:.2f}\n")
-        out.write(f"Std:    {cov_stats.global_std:.2f}\n")
-        out.write(f"Min:    {cov_stats.global_min:.2f}\n")
-        out.write(f"Max:    {cov_stats.global_max:.2f}\n\n")
-
-        out.write("Percentiles (global):\n")
-        for p in percentiles:
-            val = cov_stats.df["cov"].quantile(p/100)
-            out.write(f"  {p:>6.2f}% : {val:.2f}\n")
-
-        # now writing per-contig
-        out.write("\n========================================\n")
-        out.write("=== PER-CONTIG WINDOW COVERAGE STATS ===\n")
-        out.write("========================================\n")
-        for contig, group in cov_stats.df.groupby("contig", sort=False):
-            out.write(f"\n-- {contig} --\n")
-            contig_length = contig_lengths.get(contig, 0)
-            gm = group["cov"].mean()
-            md = group["cov"].median()
-            sd = group["cov"].std()
-            mn = group["cov"].min()
-            mx = group["cov"].max()
-            out.write(f"Contig size: {contig_length:,} bp\n")
-            out.write(f"Number of windows: {len(group)}\n")
-            out.write(f"Mean:   {gm:.2f}\n")
-            out.write(f"Median: {md:.2f}\n")
-            out.write(f"Std:    {sd:.2f}\n")
-            out.write(f"Min:    {mn:.2f}\n")
-            out.write(f"Max:    {mx:.2f}\n")
-            out.write("\nPercentiles:\n")
-            for p in percentiles:
-                val = group["cov"].quantile(p/100)
-                out.write(f"  {p:>6.2f}% : {val:.2f}\n")
-            out.write("\n")
-
-    tee(f"\n  Window-coverage summary written to:\n      {Fore.YELLOW}{out_txt}{Fore.RESET}", log_file)
-
-    # writing out as tsv
     df_summary = pd.DataFrame(rows)
     cols = ["name","length_bp","num_windows","mean","median","std","min","max"] \
            + [f"p{p}" for p in percentiles]
     df_summary.to_csv(out_tsv, sep="\t", index=False, float_format="%.2f", columns=cols)
 
-    tee(f"  Window-coverage summary table written to:\n      {Fore.YELLOW}{out_tsv}{Fore.RESET}", log_file)
+    tee(f"\n  Window-coverage overview written to:\n      {Fore.YELLOW}{out_tsv}{Fore.RESET}", log_file)
 
 
 def write_windows_table(cov_stats, output_dir, log_file):
@@ -697,10 +650,10 @@ def write_regions_of_interest_table(merged_regions, output_dir, type, log_file):
     if len(sorted_df) > 0:
         n_low_complexity = int(sorted_df["low_complexity"].sum()) if "low_complexity" in sorted_df.columns else 0
         lc_note = f" ({n_low_complexity} flagged as low-complexity)" if n_low_complexity > 0 else ""
-        tee(f"\n  Number of {type}-coverage regions identified: {Fore.YELLOW}{len(sorted_df)}{Fore.RESET}{lc_note}", log_file)
+        tee(f"\n\n  Number of {type}-coverage regions identified: {Fore.YELLOW}{len(sorted_df)}{Fore.RESET}{lc_note}", log_file)
         tee(f"    {type.capitalize()}-coverage regions-of-interest table written to:\n      {Fore.YELLOW}{out_path}{Fore.RESET}", log_file)
     else:
-        tee(f"\n  No {type}-coverage regions-of-interest identified.", log_file)
+        tee(f"\n\n  No {type}-coverage regions-of-interest identified.", log_file)
 
 
 def write_regions_fasta(reference_fasta, regions_df, buffer, output_dir, type, contig_lengths):