TY - JOUR
T1 - Precise reconstruction of the TME using bulk RNA-seq and a machine learning algorithm trained on artificial transcriptomes
AU - Zaitsev, Aleksandr
AU - Chelushkin, Maksim
AU - Dyikanov, Daniiar
AU - Cheremushkin, Ilya
AU - Shpak, Boris
AU - Nomie, Krystle
AU - Zyrin, Vladimir
AU - Nuzhdina, Ekaterina
AU - Lozinsky, Yaroslav
AU - Zotova, Anastasia
AU - Degryse, Sandrine
AU - Kotlov, Nikita
AU - Baisangurov, Artur
AU - Shatsky, Vladimir
AU - Afenteva, Daria
AU - Kuznetsov, Alexander
AU - Paul, Susan Raju
AU - Davies, Diane L.
AU - Reeves, Patrick M.
AU - Lanuti, Michael
AU - Goldberg, Michael F.
AU - Tazearslan, Cagdas
AU - Chasse, Madison
AU - Wang, Iris
AU - Abdou, Mary
AU - Aslanian, Sharon M.
AU - Andrewes, Samuel
AU - Hsieh, James J.
AU - Ramachandran, Akshaya
AU - Lyu, Yang
AU - Galkin, Ilia
AU - Svekolkin, Viktor
AU - Cerchietti, Leandro
AU - Poznansky, Mark C.
AU - Ataullakhanov, Ravshan
AU - Fowler, Nathan
AU - Bagaev, Alexander
N1 - Publisher Copyright:
© 2022 Elsevier Inc.
PY - 2022/8/8
Y1 - 2022/8/8
N2 - Cellular deconvolution algorithms virtually reconstruct tissue composition by analyzing the gene expression of complex tissues. We present the decision tree machine learning algorithm, Kassandra, trained on a broad collection of >9,400 tissue and blood sorted cell RNA profiles incorporated into millions of artificial transcriptomes to accurately reconstruct the tumor microenvironment (TME). Bioinformatics correction for technical and biological variability, aberrant cancer cell expression inclusion, and accurate quantification and normalization of transcript expression increased Kassandra stability and robustness. Performance was validated on 4,000 H&E slides and 1,000 tissues by comparison with cytometric, immunohistochemical, or single-cell RNA-seq measurements. Kassandra accurately deconvolved TME elements, showing the role of these populations in tumor pathogenesis and other biological processes. Digital TME reconstruction revealed that the presence of PD-1-positive CD8+ T cells strongly correlated with immunotherapy response and increased the predictive potential of established biomarkers, indicating that Kassandra could potentially be utilized in future clinical applications.
AB - Cellular deconvolution algorithms virtually reconstruct tissue composition by analyzing the gene expression of complex tissues. We present the decision tree machine learning algorithm, Kassandra, trained on a broad collection of >9,400 tissue and blood sorted cell RNA profiles incorporated into millions of artificial transcriptomes to accurately reconstruct the tumor microenvironment (TME). Bioinformatics correction for technical and biological variability, aberrant cancer cell expression inclusion, and accurate quantification and normalization of transcript expression increased Kassandra stability and robustness. Performance was validated on 4,000 H&E slides and 1,000 tissues by comparison with cytometric, immunohistochemical, or single-cell RNA-seq measurements. Kassandra accurately deconvolved TME elements, showing the role of these populations in tumor pathogenesis and other biological processes. Digital TME reconstruction revealed that the presence of PD-1-positive CD8+ T cells strongly correlated with immunotherapy response and increased the predictive potential of established biomarkers, indicating that Kassandra could potentially be utilized in future clinical applications.
KW - bulk RNA sequencing
KW - deconvolution
KW - tumor microenvironment
UR - http://www.scopus.com/inward/record.url?scp=85135728670&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85135728670&partnerID=8YFLogxK
U2 - 10.1016/j.ccell.2022.07.006
DO - 10.1016/j.ccell.2022.07.006
M3 - Article
C2 - 35944503
AN - SCOPUS:85135728670
SN - 1535-6108
VL - 40
SP - 879-894.e16
JO - Cancer cell
JF - Cancer cell
IS - 8
ER -