This work forms part of the PID2020-113441RB-I00 project, funded by the Spanish Research State Agency (MCIN/AEI/10.13039/501100011033). The preliminary phase of the study (for feed efficiency estimation) was funded by the SMARTER project through the Horizon 2020 research and innovation program of the European Commission (Grant Agreement No. 772787). E. Barrio benefited from a FPI predoctoral contract (PRE2021-098235) funded by the Spanish Research State Agency (MCIN/AEI/10.13039/501100011033) and by the European Social Fund (“ESF Investing in your future”)., Peer reviewed

Supplementary tables to the manuscript entitled Assessment of milk metabolites as biomarkers for predicting feed efficiency in dairy sheep, Supplemental Table 1: Grids used in the hyperparameter tuning stage for each machine learning (ML) model.
Supplemental Table 2: Summary of milk production and FE phenotypes in the 39 Assaf ewes analyzed in this work. For each animal
(Sample_ID), the type of lambing (single or twins), the means (±coefficient of variation in %) of the milk production traits: milk yield (MY),
milk protein percentage (PP), milk fat percentage (FP) and milk lactose percentage (LC) are presented. The FE parameters recorded during the
three-week evaluation period: Dry matter intake (DMI), feed efficiency indices RFI and FCR and the high, medium or low-efficiency group for
each FE-index are also presented.
Supplemental Table 3: m/z and LC retention time (RT, min) for the features identified in the raw milk metabolomic signature for the 39 Assaf ewes. The four sheets show the results of the reversed-phase liquid chromatography (C8) and hydrophilic interaction liquid chromatography (HILIC) for the positive and negative poles, respectively.
Supplemental Table 4: Selected features for residual feed intake (RFI) and feed conversion ratio (FCR) based on the Variable Importance in the Projection (VIP) values obtained in the partial least squares discriminant analysis.
Supplemental Table 5: Results of the hyperparameter tunning stage for the 100 iterations of random assignment of samples in the train and training population for all machine learning models applied for both feed efficiency (FE) metrics, residual feed intake (RFI) and feed conversion ratio (FCR). The different sheets show the results of the four ML methods used (multi-layer feedforward artificial neural network (Deep), Randon Forest (RF), extreme gradient boosting (xgboost) and Support Vector Machine (SVM) using all features (All) and those selected by VIP.
Supplemental Table 6: Candidate compounds annotated for the 41 and 26 features selected for residual feed intake (RFI) and feed conversion ratio (FCR) during the partial least squares discriminant analysis for the discrimination of the feed efficiency (FE) groups.
Supplemental Table 7: Results of metabolic pathways enrichment analysis obtained using IMPaLA software for the candidate compound matches obtained during the annotation process for the metabolites selected for residual feed intake (RFI) and feed conversion ratio (FCR)., SMARTER – SMAll RuminanTs breeding for Efficiency and Resilience 772787
European Commission, Peer reviewed, v02

Supplementary Material S1: Supplementary methods. Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P.A.S. Fonseca, G. Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz.
S1 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P.A.S. Fonseca, G. Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz.
Supplementary Table S1. Description of the experimental groups defined in the nutritional challenge experiment in the Spanish Assaf dairy ewes included in the present study.
S2 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P.A.S. Fonseca, G. Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz.
Supplementary Table S2. Statistics of the milk production and feed efficiency traits measured in the first lactation of the Assaf dairy ewes included in the nutritional challenge experiment.
S3 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega. B. Gutiérrez. Gil. P. A.S. Fonseca. Gonzalo Hervás. R. Pelayo. P.G. Toral. H. Marina. P. de Frutos. J.J. Arranz.
Supplementary Table S3. Phenotypic Values for Residual Feed Intake (RFI) and Feed Conversion Ratio (FCR) indexes and group classification for transcriptomic analyses in the Spanish Assaf dairy ewes included in the present study.
S4 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S4. Differentially expressed genes in the mammary gland transcriptome between High-FE-RFI and Low-FE-RFI sheep

S5 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S5. Differentially expressed genes in the mammary gland transcriptome between High-FE-FCR and Low-FE-FCR sheep

S6 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S6. Enrichment analysis results from the differentially expressed genes with higher expression in the low-FE-FCR Assaf dairy sheep

S7 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S7. Differentially expressed genes in the mammary gland transcriptome between High-FE-consensus and Low-FE-consensus sheep

S8 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S7. Enrichment analysis results from the differentially expressed genes with higher expression in the high-FE-consensus Spanish Assaf ewe group

S9 Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes A. Suárez-Vega, B. Gutiérrez. Gil, P. A.S. Fonseca, Gonzalo Hervás, R. Pelayo, P.G. Toral, H. Marina, P. de Frutos, J.J. Arranz
Supplementary Table S9. Enrichment analysis results from the differentially expressed genes with higher expression in the low-FE-consensus Spanish Assaf ewe group., “Supplementary material for the manuscript "Milk Transcriptome Biomarker Identification to Enhance Feed Efficiency and Reduce Nutritional Costs in Dairy Ewes”., Horizon 2020 Framework Programme SMARTER. Contract No. 772787, Peer reviewed, v01