In this study we categorize the genes, add in genes from other studies recently part of our projects, get the subtypes and find out if the genes that are specific to the study top genes, the top fold change genes for EBV and HIV in Hodgkins lymphoma, or the genes relative to the study in finding specificity to tumor mutation burden or tmb, immune checkpoint inhibitor, HIV specific, and other combinations to predict the class samples in a 3 class model for diagnosis type, then in a 5 class model of subtype cellularity. The genes specifically selected from this set scored 100% accuracy in subtype prediction of Nodular Sclerosis or Mixed Cellularity and found in 2 different gene feature sets as predictors that the NA sample was the Nodular Sclerosis as a prediction. The part 1 is added to the end of this part 2. For class type, the best performing features were not the top fold change genes but all 35 genes, or top 5 genes of the study in predicting the Hodgkins or Hodgkins with EBV but not in Hodgkins with EBV and HIV like the top 5 ranked tumor mutational burden found them in more mutations within the HIV samples.

Exploratory data analysis and summary of text datasets for prediction model.

Berikut adalah praktikum simulasi statistika mengenai pengaruh ukuran sampel (n), variabilitas data , dan jenis distribusi (Z vs t) terhadap lebar interval kepercayaan 95%. Simulasi ini dilakukan menggunakan program R dengan membandingkan kondisi parameter yang diketahui dan tidak diketahui. Hasil analisis menunjukkan bahwa peningkatan ukuran sampel dan penurunan variabilitas secara signifikan mempersempit lebar interval kepercayaan, yang mengindikasikan peningkatan presisi estimasi.

GEOG588 Lab 4 - Part B Report

GEOG588 Lab 4 Part A Report on tornado data