On clustering for cell-phenotyping in multiplex immunohistochemistry (mIHC) and multiplexed ion beam imaging (MIBI) data

Seal, Souvik; Wrobel, Julia; Johnson, Amber M.; Nemenoff, Raphael A.; Schenk, Erin L.; Bitler, Benjamin G.; Jordan, Kimberly R.; Ghosh, Debashis

doi:10.1186/s13104-022-06097-x

Cited by 8 publications

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future, we would like to study this problem with a deeper focus and pursue methodological development in this area. Finally, we would like to explore the possibility of using SMASH in the context of multiplex immunohistochemistry (mIHC) datasets [63, 64] where the goal is to identify spatially variable cell types and their interaction.…”

Section: Discussionmentioning

confidence: 99%

SMASH: Scalable Method for Analyzing Spatial Heterogeneity of genes in spatial transcriptomics data

Seal

Bitler

Ghosh

2023

Preprint

Self Cite

View full text Add to dashboard Cite

In high-throughput spatial transcriptomics (ST) studies, it is of great interest to identify the genes whose level of expression in a tissue covaries with the spatial location of cells/spots. Such genes, also known as spatially variable genes (SVGs), can provide crucial biological insights into both structural and functional characteristics of complex tissues. Existing methods for detecting SVGs either suffer from huge computational demand or significantly lack statistical power. We propose a non-parametric method termed SMASH that achieves a balance between the above two problems. We compare SMASH with other existing methods in varying simulation scenarios demonstrating its superior statistical power and robustness. We apply the method to four ST datasets from different platforms revealing interesting biological insights.

show abstract

Section: Discussionmentioning

confidence: 99%

SMASH: Scalable Method for Analyzing Spatial Heterogeneity of genes in spatial transcriptomics data

Seal

Bitler

Ghosh

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We applied our method to two real datasets, an mIHC lung cancer dataset ( Seal et al , 2022b ) from Vectra 3.0 platform and a MIBI TNBC dataset ( Keren et al , 2018 ). We also applied the traditional thresholding-based method on both the datasets.…”

Section: Real-data Analysismentioning

confidence: 99%

“…The vector of estimated values of the EQMI of all the subjects is tested for association with clinical outcomes. With the proposed method, we analyzed an mIHC lung cancer dataset ( Seal et al , 2022b ) finding that a higher co-expression of the markers, HLA-DR and CK was significantly associated with better 5-year overall survival. We analyzed a MIBI triple-negative breast cancer (TNBC) dataset ( Keren et al , 2018 ) studying the co-expression of two sets of functional markers, (a) HLA-DR, CD45RO, H3K27me3, H3K9ac and HLA-Class-1, and (b) PD1, PD-L1, Lag3 and IDO, which are also known as immuno-regulatory proteins (IRP’s).…”

Section: Introductionmentioning

confidence: 99%

MIAMI: mutual information-based analysis of multiplex imaging data

Seal

Ghosh

2022

Bioinformatics

Self Cite

View full text Add to dashboard Cite

Motivation Studying the interaction or co-expression of the proteins or markers in the tumor microenvironment (TME) of cancer subjects can be crucial in the assessment of risks, such as death or recurrence. In the conventional approach, the cells need to be declared positive or negative for a marker based on its intensity. For multiple markers, manual thresholds are required for each marker, which can become cumbersome. The performance of the subsequent analysis relies heavily on this step and thus suffers from subjectivity and lacks robustness. Results We present a new method where different marker intensities are viewed as dependent random variables, and the mutual information (MI) between them is considered to be a metric of co-expression. Estimation of the joint density, as required in the traditional form of MI, becomes increasingly challenging as the number of markers increases. We consider an alternative formulation of MI which is conceptually similar but has an efficient estimation technique for which we develop a new generalization. With the proposed method, we analyzed a lung cancer dataset finding the co-expression of the markers, HLA-DR and CK to be associated with survival. We also analyzed a triple negative breast cancer dataset finding the co-expression of the immuno-regulatory proteins, PD1, PD-L1, Lag3 and IDO, to be associated with disease recurrence. We demonstrated the robustness of our method through different simulation studies. Availability The associated R package can be found here, https://github.com/sealx017/MIAMI. Supplementary information The Supplementary Material is attached.

show abstract

“…On one hand, manual gating can be subjective. On the other hand, graph-based clustering results are prone to over-clustering and producing poor separation between clusters [23,24].…”

Section: Introductionmentioning

confidence: 99%

“…On one hand, manual gating can be subjective. On the other hand, graph-based clustering results are prone to over-clustering and producing poor separation between clusters [23,24]. The challenges described above are well recognized and there are a few methods and software developed that attempt to automate cell phenotyping for mIF images [22,[25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

GammaGateR: semi-automated marker gating for single-cell multiplexed imaging

Xiong,

Kaur,

Heiser

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Multiplexed immunofluorescence (mIF) is an emerging assay for multichannel protein imaging that can decipher cell-level spatial features in tissues. However, existing automated cell phenotyping methods, such as clustering, face challenges in achieving consistency across experiments and often require subjective evaluation. As a result, mIF analyses often revert to marker gating based on manual thresholding of raw imaging data. To address the need for an evaluable semi-automated algorithm, we developed GammaGateR, an R package for interactive marker gating designed specifically for segmented cell-level data from mIF images. Based on a novel closed-form gamma mixture model, GammaGateR provides estimates of marker-positive cell proportions and soft clustering of marker-positive cells. The model incorporates user-specified constraints that provide a consistent but slide-specific model fit. We compared GammaGateR against the newest unsupervised approach for annotating mIF data, employing two colon datasets and one ovarian cancer dataset for the evaluation. We showed that GammaGateR produces highly similar results to a silver standard established through manual annotation. Furthermore, we demonstrated its effectiveness in identifying biological signals, achieved by mapping known spatial interactions between CD68 and MUC5AC cells in the colon and by accurately predicting survival in ovarian cancer patients using the phenotype probabilities as input for machine learning methods. GammaGateR is a highly efficient tool that can improve the replicability of marker gating results, while reducing the time of manual segmentation.

show abstract

On clustering for cell-phenotyping in multiplex immunohistochemistry (mIHC) and multiplexed ion beam imaging (MIBI) data

Cited by 8 publications

References 20 publications

SMASH: Scalable Method for Analyzing Spatial Heterogeneity of genes in spatial transcriptomics data

SMASH: Scalable Method for Analyzing Spatial Heterogeneity of genes in spatial transcriptomics data

MIAMI: mutual information-based analysis of multiplex imaging data

GammaGateR: semi-automated marker gating for single-cell multiplexed imaging

Contact Info

Product

Resources

About