Skip to content

[BF][ENH] Harmonization works better now #124

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
arnaudbore merged 8 commits into from
Sep 3, 2026
Merged

[BF][ENH] Harmonization works better now #124

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
794c4b7
fix harmonization config + add error if not enough participants to ru…
arnaudbore Aug 31, 2026
bc45a77
fix lint
arnaudbore Aug 31, 2026
38532f9
revert some modifications
arnaudbore Aug 31, 2026
47afa90
Update nextflow_schema.json
arnaudbore Sep 3, 2026
9312f57
Update nextflow_schema.json
arnaudbore Sep 3, 2026
acecd51
Update nextflow_schema.json
arnaudbore Sep 3, 2026
302d97e
Update nextflow_schema.json
arnaudbore Sep 3, 2026
4b263d0
answer all Manon comments
arnaudbore Sep 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions conf/modules/harmonization.config
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,13 @@ includeConfig '../../subworkflows/nf-neuro/harmonization/nextflow.config'

process {
withName: ".*:HARMONIZATION:CLINICALCOMBAT" {
ext.regul_ref = params.harmonization_regul_ref
ext.regul_mov = params.harmonization_regul_mov
ext.nu = params.harmonization_nu
ext.tau = params.harmonization_tau
ext.limit_age = params.harmonization_limit_age
ext.ignore_sex = params.harmonization_ignore_sex
ext.ignore_handedness = params.harmonization_ignore_handedness
publishDir = [
[
path: "${params.outdir}/harmonization_models/",
Expand Down
7 changes: 7 additions & 0 deletions nextflow.config
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -199,6 +199,13 @@ params {

//**Harmonization**//
harmonization_reference = null
harmonization_regul_ref = 0
harmonization_regul_mov = -1
harmonization_nu = 5
harmonization_tau = 2
harmonization_limit_age = false
harmonization_ignore_sex = false
harmonization_ignore_handedness = false

//**Output in original space**//
output_orig_space = false
Expand Down
41 changes: 41 additions & 0 deletions nextflow_schema.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1091,6 +1091,47 @@
"type": "string",
"description": "TSV file of the reference site used to harmonize the current site (e.g. the output TSV from stats/metrics in roi).",
"help_text": "TSV file of the reference site used to harmonize the current site (e.g. the output TSV from stats/metrics in roi)."
},
"harmonization_ignore_sex": {
"type": "boolean",
"default": false,
"description": "If set, ignore the sex covariate in the data.",
Comment thread
arnaudbore marked this conversation as resolved.
Outdated
"help_text": "If set, ignore the sex covariate in the data."
Comment thread
arnaudbore marked this conversation as resolved.
Outdated
},
"harmonization_ignore_handedness": {
"type": "boolean",
"default": false,
"description": "If set, ignore the handedness covariate in the data.",
Comment thread
arnaudbore marked this conversation as resolved.
Outdated
"help_text": "If set, ignore the handedness covariate in the data."
Comment thread
arnaudbore marked this conversation as resolved.
Outdated
},
"harmonization_regul_ref": {
"type": "number",
"default": 0,
"description": "Regularization parameter for the reference site data.",
"help_text": "Regularization parameter for the reference site data."
},
"harmonization_regul_mov": {
"type": "number",
"default": -1,
"description": "Regularization parameter for the moving site data. Set to '-1' for automatic tuning [default=0 for pairwise; -1 for clinical].",
"help_text": "Regularization parameter for the moving site data. Set to '-1' for automatic tuning [default=0 for pairwise; -1 for clinical]"
},
"harmonization_nu": {
"type": "number",
"default": 5,
"description": "Combat Clinical hyperparameter for the standard deviation estimation of the moving site data. It must be >=0. [5]",
"help_text": "Combat Clinical hyperparameter for the standard deviation estimation of the moving site data. It must be >=0. [5]"
},
"harmonization_tau": {
"type": "number",
"default": 2,
"description": "Combat Clinical hyperparameter for the covariate fit of the moving site data. It must be >= 1.",
"help_text": "Combat Clinical hyperparameter for the covariate fit of the moving site data. It must be >= 1."
},
"harmonization_limit_age": {
"type": "boolean",
"description": "If set, exclude reference site subjects with age outside the range of the moving site subject ages.",
"help_text": "If set, exclude reference site subjects with age outside the range of the moving site subject ages."
}
},
"help_text": "Parameters to configure the harmonization subworkflow."
Expand Down
8 changes: 4 additions & 4 deletions subworkflows/local/utils_nfcore_sf-tractomics_pipeline/main.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -336,8 +336,8 @@ workflow PIPELINE_INITIALISATION {
readout: readout,
pe: pe_check.pe,
age: item.meta.age,
sex: item.meta.sex ? (item.meta.sex?.toLowerCase().contains("f") ? 2 : item.meta.sex.toLowerCase().contains("m") ? 1 : 0) : 0,
handedness: item.meta.handedness ? (item.meta.handedness?.toLowerCase().contains("r") ? 1 : item.meta.handedness?.toLowerCase().contains("l") ? 2 : "NA") : "NA",
sex: item.meta.sex ? (item.meta.sex.toString().trim().toLowerCase() in ['m', 'male', '1'] ? 1 : item.meta.sex.toString().trim().toLowerCase() in ['f', 'female', '2'] ? 2 : 1) : 1,
handedness: item.meta.handedness ? (item.meta.handedness?.toLowerCase().contains("r") ? 1 : item.meta.handedness?.toLowerCase().contains("l") ? 2 : 1) : 1,
disease: item.meta.disease ?: "HC",
site: item.meta.site],
t1w,
Expand Down Expand Up @@ -439,8 +439,8 @@ workflow PIPELINE_INITIALISATION {
readout: readout,
pe: pe,
age: item.meta.age,
sex: item.meta.sex ? (item.meta.sex?.toLowerCase().contains("f") ? 2 : item.meta.sex.toLowerCase().contains("m") ? 1 : 0) : 0,
handedness: item.meta.handedness ? (item.meta.handedness?.toLowerCase().contains("r") ? 1 : item.meta.handedness?.toLowerCase().contains("l") ? 2 : "NA") : "NA",
sex: item.meta.sex ? (item.meta.sex.toString().trim().toLowerCase() in ['m', 'male', '1'] ? 1 : item.meta.sex.toString().trim().toLowerCase() in ['f', 'female', '2'] ? 2 : 1) : 1,
handedness: item.meta.handedness ? (item.meta.handedness?.toLowerCase().contains("r") ? 1 : item.meta.handedness?.toLowerCase().contains("l") ? 2 : 1) : 1,
disease: item.meta.disease ?: "HC",
site: item.meta.site],
t1w,
Expand Down
36 changes: 35 additions & 1 deletion workflows/sf-tractomics.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,6 +38,7 @@ workflow SF_TRACTOMICS {

ch_inputs = ch_inputs
.multiMap{ meta, t1, wmparc, aparcaseg, dwi_bval_bvec, rev_dwi_bval_bvec, b0, rev_b0, lesion ->
meta: meta
t1: [meta, t1]
wmparc: [meta, wmparc]
aparcaseg: [meta, aparcaseg]
Expand All @@ -48,6 +49,14 @@ workflow SF_TRACTOMICS {
lesion: [meta, lesion]
}

if ( params.harmonization_reference ) {
ch_inputs.meta
.collect()
.map { meta_list ->
validateMinimumParticipantsPerSite(meta_list)
}
}

ch_versions = channel.empty()
ch_sub_multiqc_files = channel.empty()
ch_global_multiqc_files = channel.empty()
Expand Down Expand Up @@ -267,7 +276,6 @@ workflow SF_TRACTOMICS {
// while keeping the header from the first
// file only and skipping it in the rest.
ch_collection_mean_input = ATLAS_ROIMETRICS.out.stats_tab_mean

ch_collection_mean_input = collectStatsFiles(ch_collection_mean_input, "space-native_atlas-iit_label-mean_desc-roi_stats.tsv", "${params.outdir}/metrics/")
ch_global_multiqc_files = ch_global_multiqc_files.mix(ch_collection_mean_input)

Expand Down Expand Up @@ -531,6 +539,32 @@ def collectStatsFiles(ch_stats_files, name, storeDir) {
}
}

def validateMinimumParticipantsPerSite(meta_list) {
def participants_per_site = [:].withDefault { [] as Set }

meta_list.each { meta ->
def meta_site = meta.get('site')
def participant = meta.get('id')
if (meta_site && participant) {
participants_per_site[meta_site.toString().trim()].add(participant.toString().trim())
}
}

def sites_with_too_few_participants = [:]
participants_per_site.each { entry ->
if (entry.value.size() < 5) {
sites_with_too_few_participants[entry.key] = entry.value
}
}
if (sites_with_too_few_participants) {
def site_errors = []
sites_with_too_few_participants.each { entry ->
site_errors << "Site '${entry.key}' has ${entry.value.size()} participant(s); at least 5 are required."
}
error "Harmonization cannot continue:\n${site_errors.join('\n')}"
}
}

/*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
THE END
Expand Down
Loading