Chapter 1 Welcome to the Data Science Workbook

1.1 Data Science learning objectives

Introduction to Data Science is a ‘sub-module’ that will run throughout your first year as part of the BIO-4008Y anc CHE-4602Y Skills modules. Here we aim to teach you the fundamentals of how to collect and organise data in a clean and sensible format and how to correctly visualise and describe data. The platform we will be using to manage, manipulate and analyse data is posit Cloud. This is a cloud based interface that utilises the programming language R in a very similar interface to RStudio (a package you may wish to install on your own computers at some point). Over the course of the next few weeks you will start to learn fundamentals of this language and its application in posit Cloud.

1.2 Teaching layout for Data Science

In Semester 1 we will be learning some basic R syntax, how to create and load data sets and how to produce some basic data visualisations that are attractive and reproducible. Each week you will have either a lecture or a 1 hour workshop, you will also be expected to complete a chapter of this workbook each week, in your own time.

Teaching for semester 1;

  • Semester 1 Lectures - Weeks; 2, 3, 6, 7, 9, 10
  • Semester 1 Workshops - Weeks; 4, 8, 11

In semester 2 will move onto some slightly more advanced data interpretation and visualisation skills and some descriptive statistics. Towards the end of the module we will briefly touch on inferential statistics, but this will be a very brief introduction. You will have a lecture every week until week 8 and a 2 hour workshop every other week. The workshops are all outlined as chapters in this workbook, if you don’t complete a workshop in the timetabled slot, we strongly recommend you make every effort to complete it in your own time.

Teaching for semester 2;

  • Semester 2 Lectures - Weeks; 1, 2, 3, 4, 5, 6, 7, 8
  • Semester 2 Workshops - Weeks; 2, 4, 6, 8

1.3 Assessment

  • Semester 1

At the end of Semester 1 you will be given the opportunity to submit a formative assignment. From this you will get feedback and a mock grade for your work, the formative assignment and marking scheme is similar to the the later summative assignment (in Semester 2) so is a great opportunity to get constructive feedback on your work and places to improve on.

  • Semester 2

During Semester 2 you will be given an assessment data set and a series of questions, you will be asked to produce a series of plots and format them into a multi-panel figure with accompanying figure legend. You will be asked to upload these alongside your R script and these will constitute the summative task for the Data Sciences component of your respective skills module.

1.4 How to use this workbook

In semester 1 each week of taught material is accompanied by a workbook chapter. You will be expected to read through these chapters and complete any exercises in your own time. Each chapter should take no more then an hour to complete and in many cases will take much less time. I will aim to set aside 10 minutes at the end of each lecture to cover any potential problems you may be having, so if you get stuck, use this time to ask for help. In semester 2 we have more face to face workshop time, in each workshop you will be expected to work through a chapter in this workbook, you may not complete this in the allocated time, but will be expected to do so in your own time. Use face to face time to ask lots of questions, especially if you are stuck.

1.5 Why is data science important and why are we teaching it in R?

It is a common misconception, that to be a good biologist or biochemist you need to ‘know’ the mechanics of how life works. Questions like; “How does a cell undergo respiration?”, “How do kidneys filter blood?”, “How do some plants fix nitrogen?” and “How do honey bees communicate the location and quantity of resources to each other?”, spring to mind. While these questions are important, they are also, now, fairly well understood. But how did biologists come to their understanding of these mechanisms? The answer to this lies in data.

Good science is based on empirical observations and these observations should be reliably collected and reproducible. Exploration of theories through observations and experimentation leads to the collection of data and analysis and interpretation of data feeds into the bedrock of our understanding of how life works, i.e. the biological sciences.

Hopefully you can see why data handling, analysis and interpretation are important. So why are we teaching you how to handle data using R?

To many of you the use of programming languages, such as R, will be new. However if you can get into the habit of manipulating and analysing data in R you will be well set on the path to becoming an efficient and effective data analyst. Being confident with data is a key skill in the sciences and will serve you well in many career paths. In addition, knowledge and experience of programming languages such as R are fast becoming key skills in their own right, in science, industry, government and beyond.

In terms of the use of R for data handling and analysis, you may notice that the term reproducible reoccurs throughout this workbook. In the same way that your methods of data collection should be conducted and recorded in such a way that they may be reproduced by others, your data manipulation and analysis must also be conducted and recorded so as to be reproducible. Using R within the posit Cloud interface (we will come back to this in Chapter 2) makes it easy to record your data manipulation and analysis workflow and, if done well, makes it very easy for others to see and repeat what you have done.

So hopefully I have convinced you that data handling, analysis and R are all worthwhile skills to cultivate. We will spend some time in lectures discussing this as well. But for now take a look at Chapter 2. Happy coding!