From ceae8c55fed590c9f5ff9b7261aa024e47e73a91 Mon Sep 17 00:00:00 2001
From: Jennifer Chang <jennifer.chang.bioinform@gmail.com>
Date: Wed, 27 Dec 2023 12:41:00 -0800
Subject: [PATCH] Adjust input length on nextclade runs

Only run nextclade on sequences at least 1400 nt long, the approximate length
of the dengue E gene. This would avoid misclassification on short sequences.
---
 ingest/workflow/snakemake_rules/nextclade.smk | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/ingest/workflow/snakemake_rules/nextclade.smk b/ingest/workflow/snakemake_rules/nextclade.smk
index a7a73e46..2e632e29 100644
--- a/ingest/workflow/snakemake_rules/nextclade.smk
+++ b/ingest/workflow/snakemake_rules/nextclade.smk
@@ -8,6 +8,8 @@ rule nextclade_all:
     output:
         "data/nextclade_results/nextclade_all.tsv",
     threads: 4
+    params:
+        min_length=1400,  # approximately E gene length
     shell:
         """
         nextclade run \
@@ -15,6 +17,7 @@ rule nextclade_all:
           -j {threads} \
           --output-tsv {output} \
           --min-match-rate 0.01 \
+          --min-length {params.min_length} \
           --silent \
           {input.sequences}
         """