From 114f8aff91498c00c626b3ceab4f370830e06e8b Mon Sep 17 00:00:00 2001
From: Haibao Tang <tanghaibao@gmail.com>
Date: Tue, 7 Jan 2025 18:20:47 -0800
Subject: [PATCH] Dev (#739)

* update help string

* Add --tandem_Nmax

* only keep lower and upper version
---
 .github/workflows/build.yml     |  2 +-
 src/jcvi/compara/blastfilter.py |  2 +-
 src/jcvi/compara/catalog.py     | 15 +++++++++++----
 3 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index 65af5d3e..1f74df33 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -9,7 +9,7 @@ jobs:
       max-parallel: 4
       matrix:
         os: [ubuntu-latest, macos-latest]
-        python-version: ["3.8", "3.10", "3.12"]
+        python-version: ["3.8", "3.12"]
 
     steps:
       - uses: actions/checkout@v4
diff --git a/src/jcvi/compara/blastfilter.py b/src/jcvi/compara/blastfilter.py
index 6b53324e..7a4f102d 100755
--- a/src/jcvi/compara/blastfilter.py
+++ b/src/jcvi/compara/blastfilter.py
@@ -307,7 +307,7 @@ def main(args):
         type=float,
         default=0.7,
         help="retain hits that have good bitscore. a value of 0.5 means "
-        "keep all values that are 50% or greater of the best hit. "
+        "keep all values that are 50%% or greater of the best hit. "
         "higher is more stringent",
     )
     p.add_argument("--exclude", help="Remove anchors from a previous run")
diff --git a/src/jcvi/compara/catalog.py b/src/jcvi/compara/catalog.py
index fe2d4247..f17ba088 100644
--- a/src/jcvi/compara/catalog.py
+++ b/src/jcvi/compara/catalog.py
@@ -199,8 +199,6 @@ def enrich(args):
             (k, [x for x in leftover if info[x] == k]) for k in missing_taxa
         )
 
-        # print genes, leftover
-        # print leftover_sorted_by_taxa
         solutions = []
         for solution in product(*leftover_sorted_by_taxa.values()):
             score = sum(weights.get((a, b), 0) for a in solution for b in genes)
@@ -619,6 +617,12 @@ def ortholog(args):
         action="store_true",
         help="Run in full 1x1 mode, including blocks and RBH",
     )
+    p.add_argument(
+        "--tandem_Nmax",
+        type=int,
+        default=10,
+        help="merge tandem genes within distance",
+    )
     p.add_argument("--cscore", default=0.7, type=float, help="C-score cutoff")
     p.add_argument(
         "--dist", default=20, type=int, help="Extent of flanking regions to search"
@@ -714,8 +718,11 @@ def ortholog(args):
 
     filtered_last = last + ".filtered"
     if need_update(last, filtered_last, warn=True):
-        # If we are doing filtering based on another file then we don't run cscore anymore
-        dargs = [last, "--cscore={}".format(ccscore)]
+        dargs = [
+            last,
+            f"--cscore={ccscore}",
+            f"--tandem_Nmax={opts.tandem_Nmax}",
+        ]
         if exclude:
             dargs += ["--exclude={}".format(exclude)]
         if opts.no_strip_names: