Machine learning
High-dimensional omics or clinical tables yield stability-scored signature panels via nested cross-validation, commented code, and a Methods draft [1].
| Category | Details |
|---|---|
| Supported platforms / instruments | Feature matrices from bulk RNA-seq (variance-stabilized or voom-normalized tables), microarray, Olink NPX, MaxQuant/DIA-NN/MSstats protein tables, metabolomics peak tables, methylation beta values, clinical/laboratory tables; single-cell pseudobulk or embedding summaries when scoped |
| Input requirements | Sample-level matrix (samples × features); binary or multiclass outcome labels; ≥2 outcome groups; enough samples per class for stratified nested CV without empty folds. Feasibility confirmed at kickoff [4, 5]. p >> n is common |
| Reference builds supported | Human GRCh38 (GENCODE v44 / Ensembl 110); mouse GRCm39 (GENCODE vM33 / Ensembl 110); UniProt Swiss-Prot for protein features; not applicable for purely clinical tables |
| Primary tools (with versions) | scikit-learn 1.6.1; glmnet 4.1-8; Boruta 11.0.0; XGBoost 2.1.3; SHAP 0.46.0; clusterProfiler 4.20.0; limma 3.60; scikit-survival 0.24.1 when survival endpoints scoped; nestedcv (R) when client prefers R-native nested CV. Versions pinned per project |
| Typical turnaround time | 3-5 weeks (single cohort, one outcome, ≤500 features, internal nested CV); 5-8 weeks (external cohort validation, multi-omics harmonization, or survival signatures). Confirmed at kickoff |
| Deliverable formats | selected_features.csv; signature_performance_summary.csv; nested-CV prediction tables; PDF/SVG ROC, precision-recall, stability, and SHAP plots; serialized models (.pkl, .rds); DOME summary table; commented R/Python scripts with renv.lock or conda export; Methods draft |
| Key cited best-practice reference | Díaz-Uriarte et al. [1], PLOS Computational Biology; Walsh et al. [6], DOME, Nature Methods |
| Custom / bespoke analysis | Non-standard outcomes, custom feature engineering, client-specified selection methods (for example Boruta vs glmnet), multi-omics fusion, ensemble selection, or alternate deliverable formats scoped at kickoff |
Biomarker discovery uses supervised machine learning on high-dimensional feature matrices to find a small multivariate panel that separates outcome groups or ranks risk [1]. Univariate differential testing plus a top-N cutoff ranks features one at a time. Discovery instead optimizes correlated feature combinations under regularization or wrapper selection inside cross-validation [2]. Mirza et al. [3] applied seven ML methods to 701 breast cancer samples across 11 GEO datasets, yielding nine-gene diagnostic and eight-gene prognostic signatures validated by qRT-PCR. Pepkio starts from client-supplied or Pepkio-processed matrices. Custom workflows are agreed at kickoff. See the biomarker discovery glossary.
Use biomarker discovery when labeled samples provide a high-dimensional feature matrix and the question calls for a compact multivariate signature. Univariate ranking alone, or a pre-specified analyte panel, usually does not need this workflow [1].
| Approach | Best for | Limitations | Approximate cost range |
|---|---|---|---|
| ML biomarker discovery (nested CV + stability scoring) | High-dimensional omics or clinical data; compact multi-feature signatures; pilot panels for targeted follow-up | Needs labeled outcomes; small n limits signature reliability; strict leakage control required | Quote-based; bioinformatics effort scales with cohort size, feature count, and validation scope |
| Univariate DE + top-N cutoff | Exploratory ranking; large univariate effect sizes | No multivariate optimization; ignores correlation structure; unstable at high dimensionality | Lower bioinformatics cost |
| Hypothesis-driven fixed panel | Known analytes with prior clinical or mechanistic evidence | Cannot discover novel feature combinations | Assay cost dominates; minimal bioinformatics |
Pepkio confirms the prediction target, endpoint type (binary, multiclass, or survival when scoped), signature size range, and whether an independent holdout or external cohort exists [1, 6]. Confounded batch-by-outcome designs and duplicate patients are flagged before modeling.
Tools used: Custom scoping templates; DOME checklist [6]
Output: Signed scope memo with primary metrics (AUROC, AUPRC, balanced accuracy, or C-index when survival is scoped)
Sample IDs are matched between the feature matrix and outcome table. Orientation, missingness, duplicate features, and class balance are audited. Feature identifiers are mapped to GENCODE v44 gene IDs or UniProt accessions when annotation is in scope.
Tools used: Custom validation scripts; clusterProfiler 4.20.0 for ID mapping when scoped
Output: sample_manifest.csv; feature_qc_summary.csv
Missing values, near-zero-variance features, and outlier samples are summarized. Unsupervised variance filtering is documented but not applied using outcome labels. Supervised filtering before cross-validation inflates performance estimates [4].
Tools used: scikit-learn 1.6.1 (VarianceThreshold); custom R diagnostics
Output: missingness_profile.csv; variance_filter_log.csv; sample and feature QC plots (PDF/SVG)
Pepkio applies the seven guiding questions from Bernett et al. [9] before splitting to confirm preprocessing, feature selection, and tuning occur only on training folds.
Tools used: Documented leakage checklist [9]
Output: leakage_audit.md
Patient-level nested CV separates inner-loop feature selection and tuning from outer-loop performance estimation [4, 6]. Stratification preserves class proportions when sample size permits.
Tools used: scikit-learn 1.6.1 (StratifiedKFold, GridSearchCV); nestedcv (R) when client prefers R-native nested CV [5]
Output: cv_fold_assignments.csv with outer and inner fold IDs per sample
Scaling, imputation, and optional variance filtering are fit on inner training data only. Batch is regressed with limma removeBatchEffect on training folds when batch is not confounded with outcome [10].
Tools used: scikit-learn 1.6.1 (Pipeline, ColumnTransformer); limma 3.60
Output: Fold-specific preprocessed matrices logged in pipeline parameters
Feature selection runs inside inner CV folds, not on pooled train and test data [4]. Default: glmnet LASSO or elastic-net (glmnet 4.1-8); Boruta 11.0.0 when all-relevant identification is scoped; optional univariate prefilter inside inner folds only [2].
Tools used: glmnet 4.1-8; Boruta 11.0.0; scikit-learn 1.6.1
Output: Per-fold selected feature lists; inner-CV tuning metrics
A regularized logistic or elastic-net classifier is trained on inner-selected features and evaluated on outer-fold samples. XGBoost 2.1.3 is compared when sample size and scope permit [11]. Simpler baselines are reported alongside more complex models.
Tools used: scikit-learn 1.6.1; XGBoost 2.1.3; scikit-survival 0.24.1 when survival endpoints scoped
Output: signature_performance_summary.csv (AUROC, AUPRC, balanced accuracy, or C-index per outer fold)
Selection frequency, the proportion of outer folds in which each feature is selected, is computed for every candidate feature [2, 12]. A consensus signature includes features meeting a kickoff-agreed stability threshold (often ≥50-60% of outer folds).
Tools used: Custom R/Python stability scripts; scikit-learn 1.6.1
Output: selected_features.csv (feature_id, selection_frequency, mean_coefficient, annotation); stability barplot and heatmap (PDF/SVG)
Consensus features receive SHAP attribution and pathway enrichment (SHAP 0.46.0; clusterProfiler 4.20.0). Deliverables are assembled. External or holdout evaluation is reported separately when a cohort was reserved at kickoff [10].
Tools used: SHAP 0.46.0; clusterProfiler 4.20.0
Output: pathway_enrichment_signature.csv; SHAP summary plots; final deliverable bundle; Methods draft
There is no universal minimum. Nested CV still needs enough samples per class to populate folds without empty strata. Feasibility is confirmed at kickoff [4, 5]. Smaller cohorts can support exploratory work but should not claim external validity without a holdout [10].
Yes, within documented limits. High missingness, extreme imbalance, or batch confounded with outcome are flagged before modeling. Pepkio does not claim robust signatures from severely underpowered or confounded designs without a feasibility discussion at kickoff.
Pepkio accepts feature matrices from bulk RNA-seq, microarray, Olink NPX, MaxQuant/DIA-NN/MSstats protein tables, metabolomics peak tables, methylation beta values, and clinical-laboratory tables. Single-cell pseudobulk or embedding summaries are accepted when scoped. Upstream Olink QC or MS preprocessing can be scoped through related proteomics spokes. Matrices must be sample-level with documented upstream normalization.
Single-cohort projects with one binary outcome and ≤500 features often finish in 3-5 weeks from data receipt. External cohort validation, multi-omics harmonization, survival endpoints, or bespoke ensemble selection may need 5-8 weeks. Confirmed at kickoff.
Batch is diagnosed with PCA and correlation checks. When batch is not confounded with outcome, Pepkio regresses batch with limma removeBatchEffect within training folds only [10]. Confounded designs are flagged at kickoff.
Yes. You retain full ownership of all code, scripts, and results. Pepkio delivers commented R and/or Python scripts with renv.lock or conda environment exports, organized by pipeline stage with a README. Delivery is via private Git repository or agreed file transfer.
Yes. Checkpoint reviews occur after QC, leakage audit, consensus signature selection, and before final delivery. You can review outcome definitions, selection methods, stability thresholds, and figures within agreed scope.
Support covers clarification of CV design, feature selection methods, stability metrics, and minor figure or table revisions within agreed scope (typically ≤20% of deliverables). Pepkio drafts Methods and Supplementary text for analyses we performed. Substantial new outcomes, cohorts, or selection methods requested by reviewers are scoped as separate milestones.
No. Pepkio is a fee-for-service provider and does not require co-authorship unless that is discussed in advance. Acknowledgment of bioinformatics support in the Acknowledgments section is the usual practice.
LASSO or elastic-net via glmnet is the default when you need a compact, sparse signature in p >> n settings [13]. Boruta is a better fit when the goal is to identify all statistically relevant features rather than a minimal panel [14]. Pepkio compares both within nested CV when scoped and reports selection stability for each.
Pepkio reports selection frequency, the proportion of outer CV folds in which each feature is selected, and visualizes stability with barplots and feature-by-fold heatmaps [2]. The consensus signature lists features meeting a kickoff-agreed threshold (often ≥50-60% of folds). Unstable features selected in only one or two folds are documented but excluded from the primary panel unless you request otherwise.
Panel size depends on assay feasibility and cohort size, not a fixed rule. Published examples range from eight-gene panels to multi-omic predictors integrating dozens of features [3, 7]. Pepkio favors parsimony: glmnet LASSO and stability filtering tend to produce smaller panels. Trade-offs between signature size, stability, and outer-fold performance are documented in the deliverable report.
Tell us:
We'll tell you:
Contact us to start with a free consultation. Need everyday bench calculators? Try our free lab tools.
