Page 132 - GPD-3-4

Gene & Protein in Disease A pyroptosis-related gene signature in myeloma

training group and test group, randomly. In addition, data Table 1. Pyroptosis-related genes identified in this study
of 70 healthy individuals, which served as controls, were
retrieved from the GTEx database for analysis of bone Gene symbol Definition
marrow mRNA expression. AIM2 Absent in melanoma 2
APIP Apoptosis protease-activating factor 1
2.2. Analysis of differential PRGs Interacting Protein

An extensive review of the literature was performed to BAK1 BRI1-associated receptor kinase 1
identify 57 PRGs 30,31 as of December 2023, which are BAX Bcl-2-associated X protein
shown in Table 1. The “limma” package in R was utilized CASP1 Cysteine-aspartic acid protease-1
to assess the significance of differentially expressed genes CASP3 Cysteine-aspartic acid protease-3
(DEGs). The findings were illustrated using heatmap. The CASP4 Cysteine-aspartic acid protease-4
protein–protein interactions (PPIs) network of DEGs was CASP5 Cysteine-aspartic acid protease-5
analyzed using the GeneMANIA platform.
CASP6 Cysteine-aspartic acid protease-6
2.3. Establishing a prognostic risk model derived CASP8 Cysteine-aspartic acid protease-8
from DEGs CASP9 Cysteine-aspartic acid protease-9
Univariate Cox analysis was performed with a P-value CHMP2A Charged Multivesicular Body Protein 2A
threshold of 0.05. This process identified preliminary CHMP2B Charged Multivesicular Body Protein 2B
prognostic genes for further study, and their prognostic CHMP3 Charged Multivesicular Body Protein 3
significance was evaluated using consensus clustering CHMP4A Charged Multivesicular Body Protein 4A
analysis. Subsequently, the “glmnet” package in R was used CHMP4B
to conduct a stepwise Lasso-penalized Cox regression Charged Multivesicular Body Protein 4B
analysis. The risk scoring formula used in the Lasso CHMP4C Charged Multivesicular Body Protein 4C
32
33
regression analysis is referenced from the literature. The CHMP6 Charged Multivesicular Body Protein 6
“survival” and “timeROC” packages were used to perform CHMP7 Charged Multivesicular Body Protein 7
receiver operating characteristic (ROC) curve analysis CYCS Recombinant Cytochrome C, Somatic
to evaluate the effectiveness of the prognostic model. DHX9 DEAH-box helicase 9
34
Both univariate and multivariate Cox regression analyses ELANE Elastase, neutrophil expressed
were employed to explore the independent prognostic FOXO3 Forkhead box O3
value of the risk score. In addition, to visually assess the
ability of PRGs to differentiate patients, we performed GPX4 Glutathione peroxidase 4
principal component analysis (PCA) using the “limma” GSDMA Gasdermin A
and “scatterplot3d” packages in R. The same algorithm GSDMB Gasdermin B
35
was applied to construct the prognostic risk model in the GSDMC Gasdermin C
internal test group. GSDMD Gasdermin D

2.4. Construction of a prognostic nomogram GSDME Gasdermin E
GZMA Granzyme A
The “rms” and “survival” packages in R, commonly used
in medical research for prognostic modeling and survival GZMB Granzyme B
36
analysis, were utilized to create prognostic nomogram. HMGB1 High-mobility group box-1 protein
For MM patients, we combined the risk score with clinical IL18 Interleukin 18
indicators such as gender, age, and disease stage to construct IL1A Interleukin 1 a
a prognostic model for predicting patient survival time. IL1B Interleukin 1 beta
The prognostic model (nomogram) provides a graphical IL6 Interleukin 6
representation that incorporates multiple prognostic IRF1 Interferon regulatory factor 1
factors, enabling a more detailed and individualized IRF2 Interferon regulatory factor 2
estimation of survival time for MM patients.
NAIP Neuronal apoptosis inhibitory protein
2.5. Immunotherapeutic response prediction NLRC4 NLR family CARD domain containing 4
Single-sample gene set enrichment analysis (ssGSEA) NLRP1 NLR family pyrin domain containing 1
from the “GSVA” R package and the Tumor Immune NLRP2 NLR family pyrin domain containing 2
Dysfunction and Exclusion (TIDE) framework were used (Cont'd...)

Volume 3 Issue 4 (2024) 3 doi: 10.36922/gpd.4534