Data Capentry

偶遇

五月初一次偶然的机会，在外导得知我在自学R语言后，推荐了一些课程给我。其中，一个学校机构组织的Data workshop通知吸引了我，在得知这是一个公益限量面向博士开课的课程后，我还是抱着试试看的态度报了名。报名截止后，我还发邮件问了主办方，被告知接下来的一周会给回复。五月底，自己收到了两个课程的正式通知，也得知后面还有很多申请的同学在排队，庆幸自己得到了这么来之不易的学习机会，必将好好珍惜。

Introduction Data Carpentry

Data Carpentry在国外很盛行，类似一些Volunteers定期为学生或研究人员组织的培训活动。对于这次活动，是KCL几位老师为wet-lab的博士们开设的数据分析课程。首次课程针对R语言零基础又对数据分析有着迫切需求的同学，课程为40人的小班教学，有两名教师轮番讲解并现场演练。课程提供免费早中餐及课件茶点，参与的学生多为生命科学领域的学生，地点选在了知名的Guys校区医院的Seminar Room。本次课程分为两天，第一天从常用的Excel入手，介绍科研中采用Excel输入及处理数据中的问题及弊端，引入R并简单入门。第二天从如何dataframe的一些常用操作入手，然后介绍了两个常用package：tidyverse和ggplot。

Lunch

课程感受

本次课程紧凑内容丰富，的确是初学者入门R的精彩课程。由于全英语授课，加之主要基于mac进行讲解，虽然有些自学基础，但到第二天的学习仍然感觉很吃力。基于来英已有仨月的生活经历，可以跟上老师的讲解，但由于坐在后排，代码看不十分清楚，加之敲代码及快捷键的使用并不熟悉，所以后期还是有难度，需要课下及时巩固学习。全班40名同学，遇到2位疑似华人学生，但由于他们英语讲的都很流利，也没好意思汉语交流，并不确定华人身份。课堂上外国学生很踊跃，反应也很快，旁边几位男生边听课边做着自己的数据分析和PPT，佩服他们超高的效率。旁边一位小姐姐也完全跟得上老师的脚步，并给我帮助很多，课间之余也跟她聊起了科研生活，更觉得自己该多下些功夫。

课堂笔记Day 1

1. Data organization in spreadsheets

1.1 Don'ts in spreadsheets

DON'T:

• modify your raw data. Always make a copy before making any changes.
• combine multiple values in one cell (units, numbers, etc)
• never mess with your raw data: always work on a duplicate copy.
• export as a text based file (csv or txt) so that R can read it.
• make calculations. When you try to export that you will not export your formulae
• Do not color code things. Computer does not care.

1.2 Names for columns:

• do not use spaces
• use UpperCaseLikeThis
• use Underscore_to_separate_words

1.3 Dealing with missing values:

• Do not use 0, because 0 is data sometimes
• NA is the best
• blank spaces also work
• Do not combine columns

1.4 Dealing with Dates

• Use buit-in functions
• create a new column to split the year from month and day
• use the formula =YEAR (#click on the cell you want to split)
• double click on the right bottom of the cell where the little cross appears. It will apply the formula all the way down
• reconstructing the date: =DATE(cell1; cell2; cell3) and this reconstructs the date based on the year, month and day
• string format: a succession of numbers

2. Introduction to R

2.1 R and RStudio

• R allows you to handle large datasets.It has lots of 'packages'.
• R Studio is like an in-built computer to work R in a more user-friendly environment.
• Every 'window' gives you information. The upper left corner is where you can write your script: you write your instructions like a lab protocol.The Console is the window in which you can execute your commands.The upper right is the Environment. Bottom right includes files, plots, packages and help.
• Pipeline: one script after the other that takes you through all the actions that you need to do to deal with your data.

2.2 Advantages of R

• it is free
• it has thousands of functions built in- so that you don't have to do this!
• it is much user friendly than other programs
• there is a large community to ask questions (and you will get an answer!)

2.3 Tips to start using R

• Be very organised. Make sub-folders that organise your project (data, outputs, figures, scripts).
• A path shows you the way: this is a series of folders and subfolders to show you where your documents are.
• Start a New Project: always whenever you are starting something new.
• Start a new R Script: this is where you will type all of your commands- your script.

2.4 You can change the appearance in R

• Windows: Options --> global options
• Mac: In the tab 'R Studio' check 'Preferences'

2.5 Object

• <- is the assign operator. This is how we assign a value to an object.
• Shortcut: Windows/Linux: "Alt" + "-" | Mac: "Option" + "-".
• Object names: with underscores, meaningful, and do not start with a number.

2.6 Useful Commands

• getwd() : it shows you where you are in your computer, it tells you the working directory
• setwd () Set working directory
• ls() : it lists the things that are in your 'workspace'.
• rm() : removes one object. THERE IS NO WAY OF RECOVERING IT!!
• sqrt () : square root
• round () : it rounds the number to whatever number of decimals that you want/need.
• length () : it tells you the number of values in a vector.
• class () : this tells you the type of object that you are dealing with
• str () : this function tells you the structure of the object
• ? #name of function It gives you the information about that function
• print () it prints the value in the screen
• (function) it prints the value in the screen
• （#）This allows you to annotate your script
• mean () It calculates the mean of a number
• args (function) Args tells you the arguments of a function
• c () Combines in one vector
• [ ] Subsets elements from vectors. The order of the elements starts in 1.
• ! means 'opposite'

2.7 Type of Data

• character – text
• Numeric (numbers)
• integer - numbers without decimals
• double - numbers with decimals
• logical - TRUE or FALSE
• In R there is a hierarchy about these types of data: logical → numeric → character ← logical

2.8 Vectors

• This is another type of object in R.
• This is just a series of values that you put together in an object using c .

2.9 Functions

• A function is a command that executes some action in your input.
• A function has 'arguments' in it: the things you input on your function so that it gets executed with your particular parameters.

2.10 Subsetting vectors

• extract values for vector use [ ].

Conditional subsetting

• AND: &
• OR: |
• Equal to: ==
• More or equal: >=
• Less or equal: <=
• More: >
• Less: <
• %in%

2.11 Missing Data

• Missing data as NA in vector.
• na.rm = TRUE (ignore the missing data)
• ( )[!is.na（）]: extract those elements are not missing.
• na.omit(): return with incomplete removed.
• ()[complete.cases()]: return with complete.

下期预告

Starting with Data

Data Manipulation using dplyr and tidyr

Data visualization with ggplot2