diff --git a/CHANGELOG.md b/CHANGELOG.md
index 609ed74f..9a7962e0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -34,7 +34,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [#313](https://github.com/nf-core/taxprofiler/pull/304) Fix pipeline not providing error when database sheet does not have a header (♥ to @noah472 for reporting, fix by @jfy133)
 - [#330](https://github.com/nf-core/taxprofiler/pull/330) Added better tagging to allow disambiguation of Kraken2 steps of Kraken2 vs Bracken (♥ to @MajoroMask for requesting, added by @jfy133)
 - [#334](https://github.com/nf-core/taxprofiler/pull/334) Increase the memory of the FALCO process to 4GB (fix by @LilyAnderssonLee)
-- Improved meta map stability for more robust pipeline resuming (fix by @jfy133)
+- [#332](https://github.com/nf-core/taxprofiler/pull/332) Improved meta map stability for more robust pipeline resuming (fix by @jfy133)
+- [#338](https://github.com/nf-core/taxprofiler/pull/338) Fixed wrong file 'out' file going to `centrifuge kreport` module (♥ to @LilyAnderssonLee for reporting, fix by @jfy133)
 
 ### `Dependencies`
 
@@ -46,6 +47,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### `Deprecated`
 
+- [#338](https://github.com/nf-core/taxprofiler/pull/338) Updated Centrifuge module to not generate (undocumented) SAM alignments by default if --save_centrifuge_reads supplied, to due to a Centrifuge bug modifying profile header. SAM alignments can still be generated if `--out-fmt` supplied in `database.csv` (♥ to @LilyAnderssonLee for reporting, fix by @jfy133)
+
 ## v1.0.1 - Dodgy Dachshund Patch [2023-05-15]
 
 ### `Added`
diff --git a/conf/modules.config b/conf/modules.config
index 8a290a6f..10f7bfac 100644
--- a/conf/modules.config
+++ b/conf/modules.config
@@ -559,7 +559,7 @@ process {
         publishDir = [
             path: { "${params.outdir}/centrifuge/${meta.db_name}/" },
             mode: params.publish_dir_mode,
-            pattern: '*.{txt,sam,gz}'
+            pattern: '*.{txt,sam,tab,gz}'
         ]
     }
 
diff --git a/modules.json b/modules.json
index cd870598..d5379c1e 100644
--- a/modules.json
+++ b/modules.json
@@ -42,7 +42,7 @@
                     },
                     "centrifuge/centrifuge": {
                         "branch": "master",
-                        "git_sha": "c8e35eb2055c099720a75538d1b8adb3fb5a464c",
+                        "git_sha": "7b0b4276fbad744a69c35d3cbf9d55fdfc9491bd",
                         "installed_by": ["modules"]
                     },
                     "centrifuge/kreport": {
diff --git a/modules/nf-core/centrifuge/centrifuge/main.nf b/modules/nf-core/centrifuge/centrifuge/main.nf
index 26c91ee1..6e45ba4e 100644
--- a/modules/nf-core/centrifuge/centrifuge/main.nf
+++ b/modules/nf-core/centrifuge/centrifuge/main.nf
@@ -5,19 +5,18 @@ process CENTRIFUGE_CENTRIFUGE {
     conda "bioconda::centrifuge=1.0.4_beta"
     container "${ workflow.containerEngine == 'singularity' && !task.ext.singularity_pull_docker_container ?
         'https://depot.galaxyproject.org/singularity/centrifuge:1.0.4_beta--h9a82719_6' :
-        'quay.io/biocontainers/centrifuge:1.0.4_beta--h9a82719_6' }"
+        'biocontainers/centrifuge:1.0.4_beta--h9a82719_6' }"
 
     input:
     tuple val(meta), path(reads)
     path db
     val save_unaligned
     val save_aligned
-    val sam_format
 
     output:
     tuple val(meta), path('*report.txt')                 , emit: report
     tuple val(meta), path('*results.txt')                , emit: results
-    tuple val(meta), path('*.sam')                       , optional: true, emit: sam
+    tuple val(meta), path('*.{sam,tab}')                 , optional: true, emit: sam
     tuple val(meta), path('*.mapped.fastq{,.1,.2}.gz')   , optional: true, emit: fastq_mapped
     tuple val(meta), path('*.unmapped.fastq{,.1,.2}.gz') , optional: true, emit: fastq_unmapped
     path "versions.yml"                                  , emit: versions
@@ -38,7 +37,6 @@ process CENTRIFUGE_CENTRIFUGE {
         unaligned = save_unaligned ? "--un-conc-gz ${prefix}.unmapped.fastq.gz" : ''
         aligned = save_aligned ? "--al-conc-gz ${prefix}.mapped.fastq.gz" : ''
     }
-    def sam_output = sam_format ? "--out-fmt 'sam'" : ''
     """
     ## we add "-no-name ._" to ensure silly Mac OSX metafiles files aren't included
     db_name=`find -L ${db} -name "*.1.cf" -not -name "._*"  | sed 's/\\.1.cf\$//'`
@@ -50,7 +48,6 @@ process CENTRIFUGE_CENTRIFUGE {
         -S ${prefix}.results.txt \\
         $unaligned \\
         $aligned \\
-        $sam_output \\
         $args
 
     cat <<-END_VERSIONS > versions.yml
diff --git a/modules/nf-core/centrifuge/centrifuge/meta.yml b/modules/nf-core/centrifuge/centrifuge/meta.yml
index a252c00c..b51e2522 100644
--- a/modules/nf-core/centrifuge/centrifuge/meta.yml
+++ b/modules/nf-core/centrifuge/centrifuge/meta.yml
@@ -27,10 +27,10 @@ input:
       type: directory
       description: Path to directory containing centrifuge database files
   - save_unaligned:
-      type: value
+      type: boolean
       description: If true unmapped fastq files are saved
   - save_aligned:
-      type: value
+      type: boolean
       description: If true mapped fastq files are saved
 output:
   - meta:
@@ -48,6 +48,11 @@ output:
       description: |
         File containing classification results
       pattern: "*.{results.txt}"
+  - sam:
+      type: file
+      description: |
+        Optional output file containing read alignments (SAM format )or a table of per-read hit information (TAB)s
+      pattern: "*.{sam,tab}"
   - fastq_unmapped:
       type: file
       description: Unmapped fastq files
diff --git a/subworkflows/local/profiling.nf b/subworkflows/local/profiling.nf
index 9e0f0165..236ed8df 100644
--- a/subworkflows/local/profiling.nf
+++ b/subworkflows/local/profiling.nf
@@ -227,7 +227,7 @@ workflow PROFILING {
                                         db: it[3]
                                 }
 
-        CENTRIFUGE_CENTRIFUGE ( ch_input_for_centrifuge.reads, ch_input_for_centrifuge.db, params.centrifuge_save_reads, params.centrifuge_save_reads, params.centrifuge_save_reads  )
+        CENTRIFUGE_CENTRIFUGE ( ch_input_for_centrifuge.reads, ch_input_for_centrifuge.db, params.centrifuge_save_reads, params.centrifuge_save_reads  )
         ch_versions            = ch_versions.mix( CENTRIFUGE_CENTRIFUGE.out.versions.first() )
         ch_raw_classifications = ch_raw_classifications.mix( CENTRIFUGE_CENTRIFUGE.out.results )
 
@@ -236,7 +236,11 @@ workflow PROFILING {
                                                 .filter { meta, db -> meta.tool == 'centrifuge' }
                                                 .map { meta, db -> [meta.db_name, meta, db] }
 
-        ch_input_for_centrifuge_kreport = CENTRIFUGE_CENTRIFUGE.out.report
+        // We must combine the _results_ file to get correct output - sending the report file will
+        // weirdly still produce valid-looking output, however the numbers are nonsense.
+        // Unfortunately the Centrifuge documentation for this was unclear as to _which_ outfile
+        // goes into it.
+        ch_input_for_centrifuge_kreport = CENTRIFUGE_CENTRIFUGE.out.results
                                             .map { meta, profile -> [meta.db_name, meta, profile] }
                                             .combine(ch_database_for_centrifugekreport, by: 0)
                                             .multiMap {