##################2019年1月18日14:34:07##########################
example("pheatmap") #獲取函數(shù)的示例
help.search("heatmap") #根據(jù)關(guān)鍵詞搜索相關(guān)的函數(shù)
library(help="pheatmap") #查看包的詳細信息
ls() #We can see objects we’ve created in the global environment
length() #return the length of vector
Alt - on Windows 快捷生成 “<-”
特點
- R does not have a type for a single value (known as a scalar) such as 3.1 or “AGCTACGACT.” Rather, these values are stored in a vector of length 1.
（R沒有類型的變量用來存儲一個值乙漓，例如字符串xx，相對應(yīng)，這些值被存儲在長度為1的向量中） - R’s vectors are the basis of one of R’s most important features: vectorization. Vectorization allows us to loop over vectors elementwise, without the need to write an explicit loop.
（向量的一個重要特點是能夠?qū)υ剡M行迭代而不需要明確的循環(huán)）
################2019年1月22日09:48:01#######################
- When we assign a value in our R session, we’re assigning it to an environment known
as the global environment. - Calling the function search() returns where R looks when searching for the value of a variable—which
includes the global environment (.GlobalEnv) and attached packages.
(當使用search()查找變量的值時，會返回R在全局變量(.GlobalEnv）以及相應(yīng)的包中查找的結(jié)果寸士。 - if one vector is longer than the other, R will recycle the values in the
shorter vector. This is an intentional behavior, so R won’t warn you when this hap‐
pens
> x <- c(1,2,3)
> x + 1
[1] 2 3 4
> y <- c(1,2)
> x + y #當兩個元素的向量不是乘積倍的時候
[1] 2 4 4
Warning message:
In x + y : longer object length is not a multiple of shorter object length
- R will return a missing value (NA; more on this later) if you try to access an ele‐
ment in a position that’s greater than the number of elements.
> z[c(2, 1, 10)]
[1] 2.2 3.4 NA
It’s also possible to exclude certain elements from lists using negative indexes
（使用負號來跳過數(shù)據(jù)）
> order(z)
[1] 4 3 5 2 1
> z[order(z)]
> order(z, decreasing=TRUE)
[1] 1 2 5 3 4
> z[order(z, decreasing=TRUE)] #order返回排序后的索引
[1] 3.4 2.2 1.2 0.4 -0.4
> sort(b,decreasing = T) #返回排序后的值
b a1 a3 a2 c
5.4 3.4 2.0 1.0 0.4
Again, often we use functions to generate indexing vectors for us. For example, one
way to resample a vector (with replacement) is to randomly sample its indexes using
the sample() function:
[1] http://www.reibang.com/p/38d0a44630f8
[2] https://bbs.pinggu.org/thread-3068145-1-1.html
> set.seed(0) # we set the random number seed so this example is reproducible
> i <- sample(length(z), replace=TRUE) #replace是否放回取樣
> i
[1] 5 2 2 3 5
> z[i]
[1] 1.2 2.2 2.2 0.4 1.2
NA is R’s built-in value to represent missing data.
NULL represents not having a value
-Inf, Inf These are just as they sound, negative infinite and positive infinite values.
NaN stands for “not a number,” which can occur in some computations that don’t
return numbers, i.e., 0/0 or Inf + -Inf.
> is.nan(0/0)
[1] TRUE
> x <- c()
> is.null(x)
[1] TRUE
> y <- c(1,2,3)
> is.na(y[4])
[1] TRUE
Because all elements in a vector must have homogeneous data type, R will silently coerce elements so that they have the same type.
(當構(gòu)建向量時，R會自動進行數(shù)據(jù)類的強轉(zhuǎn)历涝。）
- When called on numeric values, summary() returns a numeric summary with the
quartiles and the mean. - Likewise, R’s data-reading functions can also read gzipped files directly—there’s
no need to uncompress gzipped files first. - reshape2 package provides functions to reshape data: the function melt()
turns wide data into long data, and cast() turns long data into wide data. - One nice feature of data.frame() is that if you provide vectors as named arguments, data.frame() will use these names as column names.
################2019年1月23日09:29:13#######################
Omitting the row index retrieves all rows, and omitting the column index retrieves all columns.
（省略列索引將檢索所有的行兰迫，省略行索引將檢索所有的列。）
> y <- cbind(x1 = 3, x2 = c(4:1))
> y
x1 x2
[1,] 3 4
[2,] 3 3
[3,] 3 2
[4,] 3 1
> y['x1']
[1] NA
> y[1,'x1']
x1
3
> y[,'x1']
[1] 3 3 3 3
- It’s a good idea to avoid referring to specific dataframe rows in your
analysis code. - From summary(), we see that this varies quite considerably across all windows on chromosome 20:
> summary(d$total.SNPs)
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.000 3.000 7.000 8.906 12.000 93.000
- Remember, columns of a dataframe are just vectors. If you only need the data from
one column, just subset it as you would a vector: - Note that there’s no need to use a comma in the bracket
because d$percent is a vector, not a two-dimensional dataframe
> d$percent.GC[d$Pi > 16]
[1] 39.1391 38.0380 36.8368 36.7367 43.0430 41.1411 [...]
Thus, d[$Pi > 3, ] is identical to d[which(d$Pi > 3), ];
> d$Pi > 3
[1] FALSE TRUE FALSE TRUE TRUE TRUE [...]
> which(d$Pi > 3)
[1] 2 4 5 6 7 10 [...]
subset() takes two arguments: the dataframe to operate on, and then conditions to include a
row. With subset(), d[dpercent.GC > 80, ] can be expressed as:
$ subset(d, Pi > 16 & percent.GC > 80)
start end total.SNPs total.Bases depth [...]
58550 63097001 63098000 5 947 2.39 [...]
- Note that we (somewhat magically) don’t need to quote column names. This is
because subset() follows special evaluation rules, and for this reason, subset() is
best used only for interactive work.
> subset(d, Pi > 16 & percent.GC > 80,
c(start, end, Pi, percent.GC, depth))
start end Pi percent.GC depth
58550 63097001 63098000 41.172 82.0821 2.39
58641 63188001 63189000 16.436 82.3824 3.21
58642 63189001 63190000 41.099 80.5806 1.89
#####################ggplot2##################
- ggplot2 works exclusively with dataframes, so you’ll need to get your data tidy and into a dataframe before visualizing it with ggplot2.
- Each layer updates our plot by adding geometric objects such as the points in a scatterplot, or the lines in a line plot.
Geom = Geometric =幾何學
aes =aesthetic = 美學的 - We specify the mapping of aesthetic attributes to columns in our dataframe using the function aes().
