Biology subjects

Carcillo, J. A.

Publications and source records attributed to Carcillo, J. A..

3 recordsLinked to original sources

bioRxiv2025

DeepDiff-SHAP: Interpretable deep learning for subgroup-specific causal inference using conditional SHAP

Precision medicine aims to tailor healthcare strategies to individual differences in genetic, clinical, and environmental factors. However, identifying subgroup-specific causal relationships in complex biomedical data remains a major challenge, especially when standard causal inference methods average over population heterogeneity. We introduce DeepDiff-SHAP, a novel framework that combines regression-based and deep learning-based differential causal inference to detect changes in causal relationships across patient subgroups. DeepDiff-SHAP integrates SHapley Additive exPlanations (SHAP) to estimate conditional dependencies and perform nonlinear differential causal inference in a principled, interpretable manner. Applying DeepDiff-SHAP to two population-scale datasets, the CDC Diabetes Health Indicators Dataset and a UK Biobank sepsis cohort stratified by hypertension, we identified clinically meaningful and subgroup-specific causal changes in relationships between features such as age, general health, alkaline phosphatase, and cholesterol. Our results demonstrate that deep learning enhances sensitivity to complex interaction patterns overlooked by linear models, providing new biological insights into disease progression and comorbidity-specific risk mechanisms. DeepDiff-SHAP offers a scalable and interpretable solution to uncover individualized causal pathways, advancing the goal of truly personalized medicine.

bioinformatics↗

bioRxiv2024

Proteomic profiling of the local and systemic immune response to pediatric respiratory viral infections

Viral lower respiratory tract infection (vLRTI) is a leading cause of hospitalization and death in children worldwide. Despite this, no studies have employed proteomics to characterize host immune responses to severe pediatric vLRTI in both the lower airway and systemic circulation. To address this gap, gain insights into vLRTI pathophysiology, and test a novel diagnostic approach, we assayed 1,305 proteins in tracheal aspirate (TA) and plasma from 62 critically ill children using SomaScan. We performed differential expression (DE) and pathway analyses comparing vLRTI (n=40) to controls with non-infectious acute respiratory failure (n=22), developed a diagnostic classifier using LASSO regression, and analyzed matched TA and plasma samples. We further investigated the impact of viral load and bacterial coinfection on the proteome. The TA signature of vLRTI was characterized by 200 DE proteins (Padj<0.05) with upregulation of interferons and T cell responses and downregulation of inflammation-modulating proteins including FABP and MIP-5. A nine-protein TA classifier achieved an AUC of 0.96 (95% CI 0.90-1.00) for identifying vLRTI. In plasma, the host response to vLRTI was more muted with 56 DE proteins. Correlation between TA and plasma was limited, although ISG15 was elevated in both compartments. In bacterial coinfection, we observed increases in the TNF-stimulated protein TSG-6, as well as CRP, and interferon-related proteins. Viral load correlated positively with interferon signaling and negatively with neutrophil-activation pathways. Taken together, our study provides fresh insight into the lower airway and systemic proteome of severe pediatric vLRTI, and identifies novel protein biomarkers with diagnostic potential. IMPORTANCEWe describe the first proteomic profiling of the lower airway and blood in critically ill children with severe viral lower respiratory tract infection (vLRTI). From tracheal aspirate (TA), we defined a proteomic signature of vLRTI characterized by increased expression of interferon signaling proteins and decreased expression of proteins involved in immune modulation including FABP and MIP-5. Using machine learning, we developed a parsimonious diagnostic classifier that distinguished vLRTI from non-infectious respiratory failure with high accuracy. Comparative analysis of paired TA and plasma specimens demonstrated limited concordance, although the interferon-stimulated protein ISG15 was significantly upregulated with vLRTI in both compartments. We further identified TSG-6 and CRP as airway biomarkers of bacterial-viral coinfection, and viral load analyses demonstrated positive correlation with interferon-related protein expression and negative correlation with the expression of neutrophil activation proteins. Taken together, our study provides new insight into the lower airway and systemic proteome of severe pediatric vLRTI.

immunology↗

bioRxiv2021

Causal inference using deep-learning variable selection identifies and incorporates direct and indirect causalities in complex biological systems

In complex diseases, causal structure learning across biological variables is critical to identify modifiable triggers or potential therapeutic agents. A limitation of existing causal learning methods is that they cannot identify indirect causal relations, those that would interact through latent mediating variables. We developed the first computational method that identifies both direct and indirect causalities, causal inference using deep-learning variable-selection (causalDeepVASE). To accurately identify indirect causalities and incorporate them with direct causalities, causalDeepVASE develops a deep neural network approach and extends a flexible causal inference method. In simulated and biological data of various contexts, causalDeepVASE outperforms existing methods in identifying expected or validated causal relations. Further, causalDeepVASE facilitates a systematic understanding of complex diseases. For example, causalDeepVASE uniquely identified a possible causal relation between IFN{gamma} and creatinine suggested in a polymicrobial sepsis model. In future biomedical studies, causalDeepVASE can facilitate the identification of driver genes and therapeutic agents.

bioinformatics↗