as(x, "transactions")
#x为列表或data.frame

好像可以这样，试下看。

我试了这个样子不行，假设数据集是这个样子：
   cheese    bread     apple
1      0           1          1
2       1          1         0
我的输出是所有的列名，并没有按照0，1来构造准确的交易记录，是为什么呢？输出结果是：1.{cheese，bread，apple}
抱歉二次修改，之前发现自己理解错了。

# Data Processing Of Gene

# Import the dataset
library(data.table)
dataCSV = fread("Data.csv")
dataRow = fread("GSE115469_Data.csv",header = FALSE)
dataP1 = dataRow[,1:1069]

# Transpose
dataTrans = t(dataP1)

# Set the rowNames and colNames
rownames(dataTrans) = dataTrans[,1]
colnames(dataTrans) = dataTrans[1,]
dataTrans = dataTrans[-1,-1]

# character --> Numberic
dataNum=apply(dataTrans,2,as.numeric)

# Normalization
dataNum[dataNum > 1] = 1
dataNum[dataNum < 1] = 0

# Removing all rows which all columns have value
dataValue = dataNum[which(rowSums(dataNum==0)!=0),]

# Removing all rows which all columns are 0
dataZero = dataValue[which(rowSums(dataValue)>0),]
dataZero =na.omit(dataZero)

# Apriori
library(Matrix)
library(arules)

# data --> Transaction
trans = as(dataZero,"transactions")

# Using Apriori to find frequent itemsets
#dataset = read.transactions(dataTrans)
#itemFrequencyPlot(dataset, topN = 100)
rules = apriori(data = trans, parameter = list(support = 0.8, confidence = 0.8))

遇到了同样的处理Transactions交易数集的问题，是与这个基因数集格式相同的嘛，求分享