Byrd, D. R. | Search bioRxiv

bioRxiv2021

A faithful in vivo model of human macrophages in metastatic melanoma

AO_SCPLOWBSTRACTC_SCPLOWDespite recent therapeutic progress, advanced melanoma remains lethal for many patients. The composition of the immune tumor microenvironment (TME) has decisive impacts on therapy response and disease outcome. High dimensional analyses of patient samples can reveal the composition and heterogeneity of the immune TME. In particular, macrophages are known for their cancer-supportive role, but the underlying mechanisms are incompletely understood, and experimental in vivo systems are needed to test the functional properties of these cells. We characterized a humanized mouse model, reconstituted with a human immune system and a human melanoma, in which: (1) human macrophages support metastatic spread of the tumor; and (2) tumor-infiltrating macrophages have a specific transcriptional signature that faithfully represents the transcriptome of macrophages from patient melanoma samples and is associated with shorter survival. This model complements patient sample analyses, enabling the elucidation of fundamental principles in melanoma biology, and the development and evaluation of candidate therapies.

immunology↗

bioRxiv2019

Autophagy dark genes: Can we find them with machine learning?

Identifying novel genes associated with autophagy (ATG) in man remains an important task for gaining complete understanding on this fundamental physiological process. A machine-learning guided approach can highlight potentially \"missing pieces\" linking core autophagy genes with understudied, \"dark\" genes that can help us gain deeper insight into these processes. In this study, we used a set of 103 (out of 288 genes from the Autophagy Database, ATGdb), based on the presence of ATG-associated terms annotated from 3 secondary sources: GO (gene ontology), KEGG pathway and UniProt keywords, respectively. We regarded these as additional confirmation for their importance in ATG. As negative labels, we used the OMIM list of genes associated with monogenic diseases (after excluding the 288 ATG-associated genes). Data associated with these genes from 17 different public sources were compiled and used to derive a Meta Path/XGBoost (MPxgb) machine learning model trained to distinguish ATG and non-ATG genes (10-fold cross-validated, 100-times randomized models, median AUC = 0.994 +/- 0.0084). Sixteen ATG-relevant variables explain 64% of the total model gain, and 23% of the top 251 predicted genes are annotated in ATGdb. Another 15 genes have potential ATG associations, whereas 193 do not. We suggest that some of these 193 genes may represent \"autophagy dark genes\", and argue that machine learning can be used to guide autophagy research in order to gain a more complete functional and pathway annotation of this complex process.

bioinformatics↗