Introduction to scientific computing

4 Oct 2021, 09:00 → 25 Oct 2021, 18:30 Europe/Brussels

Guysen (GIGA meeting rooms)

Course presentation

Summary

The modern world is ever more dependent on computer science and information technology advances. Especially in life science in recent years, exponentially growing amounts of biological information have been obtained and deposited in various databases. The predominant source – but not only – of this data is “high-throughput” experimentation, involving simultaneous execution of hundreds or thousands of experiments. A comprehensive understanding of biological phenomena can be achieved only through the integration of all available biological information and different data analysis tools and applications. It becomes crucial for modern scientists to acquire skills that will enable them to efficiently deal with this “data explosion”. It is not surprising that scientists with strong computer skills fare better in their position and research. This course aims to provide learners with the basic notions underlying computer science as necessary for biomedical science. The focus of this course in on computer architecture, scientific computing and basic programming.

Aim of the course

To provide participants with basic knowledge on computer architecture, available storage and computing resources, good IT practices, and some programming skills in MATLAB.

By the end of the course, the participants should:

• understand personal computer & cluster architecture along with mass-storage, and their usage;
• be acquainted with the available IT resources and people incharge at the GIGA/ULiège IT infrastructure;
• be aware some principles of good programming practice;
• know about the main operational systems & programming languages;
• write basic scripts and functions in MATLAB.

---

Matlab/Octave programming

Teaching  resources

This will be the main MOOC about MATLAB programming:

• MOOC: MATLAB and Octave for Beginners;
• the whole pdf book for the course (chapters to read and exercises) will be made available or it can be purchased on Amazon.

Schedule

• On the 1st, 2nd and 3rd afternoon, we will broadly introduce the topics to be studied on the MOOC and in the book chapters (PDF files to be provided on site) for the following week. A list of exercises to be completed in pairs will also be provided.
• On the 2nd, 3rd and 4th afternoon, we will debrief the online course, answer the questions regarding the theory, and solve the exercises that created most difficulties.

The list of chapters to read, videos to watch and exercises are provided in the main time table.

---

Practicalities

Target group

PhD candidate, postdoctoral researchers and PI’s. 
Course is limited to 15 participants.

Prerequisites

For the practical and programming sessions:

• a personal laptop - with MATLAB (see the UDI-GIGA and how to contact them) or Octave installed.
• access to GIGA/SeGI resources, such as the mass-storage, Dox and the ULiege Gitlab server.

The bioinformatics team wiki provides practical information about the mass-storage and cluster, including good practice recommendation and how to connect. Don't hesitate to contact them if you don't have access or are not sure you have access to the GIGA mass storage.

Duration of the course and workload

4 days: October Tuesday 5, then Monday 11, 18, and 25.
In total, 32 hours + online classes.

Organization of the course

The course uses “blended teaching”, i.e. relies on a mixture of in-person lectures & meetings and web-based classes:

• 4 days, over the course of 4 weeks, for in-person teaching with
  • in the morning, background and more theoretical lectures;
  • in the afternoon, explanations and demo-ing for the practical parts.
• the remaining 3 weeks, with on line programming courses and exercises to solve, plus extra-lectures on topics covered in-person (when available).

Some of the lectures were recorded in October 2020 and are available here.

Speakers & Contributors

in alphabetical order:

• Mohamed Bahri
• Martin Grignard
• Gregory Hammad
• Arnaud Lavergne
• Alice Mayer
• Christophe Phillips
• Pierre-François Pirlet
• David Stern
• Yves Wesche

---

Support resources

Additional resources & extra lectures

Scientific computing, IT & Linux:

• MOOC: IT support: Hardware essentials
• MOOC: Introduction to Linux
• MOOC: Data Science: Productivity tools

Reproducibility & reliability, QA & QC in software, and Open Source Software

• lectures by Kirsite Whitaker here
• The Turing Way
• Reproducible Research
• The cost of reproducibility by Russ Poldrack

Git, GitHub, Gitlab

• a 3h lecture by Elizabeth Dupre
• Software carpentry on Git
• Mozzila on Git
• more stuff on Git by Kirstie Whitaker
• a Git Training Workshop program

 

Tuesday, 5 October

09:30 → 10:00 Welcome + Introduction to computer science : historical perspective, computer structure, operating systems & languages

Speaker: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging)

Day1_IntroComputerScience.pdf

Day1_IntroGIGADScourse.pdf

10:00 → 11:00 Data representation & storage: bits/bytes, data format, signal discretization, compression

Speaker: Dr Mohamed Bahri

Day1_DataRepresentation_Storage_2022.pdf

11:00 → 11:30 Coffee break

11:30 → 12:00 ULiège IT resources by the SeGI/UDI’s

Speaker: Mr Yves Wesche

GIGA Doctorate School_20211005.pdf

12:00 → 12:30 GIGA-ULiege Mass storage

Speaker: David Stern

20211005_MassStorage.pdf

12:30 → 13:30 Lunch break

13:30 → 14:30 Access to and use of the mass storage

Speaker: Dr Alice Mayer

2021_1005_massstorage_practical.pdf

14:30 → 15:00 Coffee break

15:00 → 17:00 Matlab & Octave practical

See the 'minutes' for the program

Speakers: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging) , Dr Mohamed Bahri

Day1_MatlabOctave_courseintro.pdf

Home-work for next week

Chapters to read:

• 1 to 5, all sections
• [6, sections 1/2 (all) + 4 (advanced) + 3/5/6 (optional)] not for beginners :-)

Video:

• 1.1-3; 2.1-5; 3.1; 4.1-3

Exercises:

• Chapter 1:
  • #9, 10, 25 (very basic)
  • #35 (interesting & useful but easier after reading Chap. 2)
• Chapter 2: #35, 42, 47 (useful vector manipulation)
• Chapter 3: #12, 29 (1st script and function!) 
• Chapter 4: #17, 18, 30 (of IF's and SWITCH's)
• Chapter 5:
  • #10, 17, 20, 32 (of FOR's and WHILE's)
  • #40 (more advanced and complicated)
• Not for beginners. :-)
  Chapter 6:  Write functions that return
  • the day index since January 1, 2021 based on a date expressed as [dd mm yyyy] (you can assume the date is taken in 2021 to begin with).
  • the name day (Monday, Tuesday,...) for a date,  expressed as [dd mm yyyy], knowing that January 1, 2021, is a Friday (you can assume the date is taken in 2021 to begin with).

Monday, 11 October

09:30 → 10:30 Introduction to algorithmic thinking: Different types of algorithms, computer logic.

Speaker: Dr Mohamed Bahri

Algorithic_GIGA_Doctoral_school_20211012.pdf

10:30 → 11:00 Coffee break

11:00 → 12:00 Code versioning & Git, an introduction.

Speaker: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging)

Day2_VersioningGit.pdf

12:00 → 13:00 Lunch break

13:00 → 15:00 GitHub/Gitlab account hands-on check and explanations

Speakers: Dr Grégory Hammad (GIGA-CRC in vivo imaging) , Martin Grignard

HandsOnTutorial_Git.pdf

15:00 → 15:30 Coffee break

15:30 → 17:30 Matlab & Octave practical

See the 'minutes' for the program

Speakers: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging) , Dr Mohamed Bahri

Home-work for next week

Chapters to read:

• Chapter 6, sections 1/2 (all) + 4 (advanced) + 3/5/6 (optional) not for beginners :-)
• Chapter 7: sections 1/2/3 + 4 (optional)
• Chapter 8: sections 1/2 + 3/4/5 (advanced)

Video:

• 3.2-4; 5.1, 5.2

Exercises:

• Chapter 6:  Write functions that return
  • the day index since January 1, 2021 based on a date expressed as [dd mm yyyy] (you can assume the date is taken in 2021 to begin with).
  • the name day (Monday, Tuesday,...) for a date,  expressed as [dd mm yyyy], knowing that January 1, 2021, is a Friday (you can assume the date is taken in 2021 to begin with).
• Chapter 7 : 6, 14, 24, 30
• Chapter 8 : 8, 26, 31

Monday, 18 October

09:30 → 10:30 Good practices in scientific computing: Code modularity, validation, optimization, documentation

Speaker: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging)

Day3_SciProgrammingGoodPractices.pdf

10:30 → 11:00 Coffee break

11:00 → 13:00 Working with High Performance Computing Clusters & parallel processing (including hands-on on GIGA cluster).

Speaker: Dr Alice Mayer

2021_clusters_and_parallel_computing.pdf

2021_clusters_and_parallel_computing.pdf

13:00 → 14:00 Lunch break

14:00 → 15:00 Research Data Management and Reproducibility: good habits for good research

Speaker: Dr Judith Biernaux (ARD ULiege)

20211018_RDM_Giga_JBiernaux.pdf

15:00 → 15:30 Coffee break

15:30 → 17:30 Matlab & Octave practical

See the 'minutes' for the program

Speakers: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging) , Dr Mohamed Bahri

Home-work for next week

Chapters to read:

• Chapter 9: sections 1 + 3 (advanced)
• Chapter 10: sections 1, 2
• Chapter 12: sections 15 + 2 (advanced)

Video:

• 3.2-4; 5.1, 5.2

Exercises:

• Chapter 10 : 2, 4, 12, 25, 26 
• Chapter 12 : 3, 12, 16

Monday, 25 October

09:30 → 10:30 Working with personal data ? Think GDPR!

Speaker: Mr Pierre-François Pirlet

20211025-Protection_des_DCP__lois_et_mesures_concrètes__GIGA_Doctoral_School_anglais.pdf

10:30 → 11:00 Coffee break

11:00 → 12:00 Data processing in genomics, from the sequencers to the results: data management, pipelines and containers, reproductibility.

Speaker: Dr Arnaud Lavergne

DATA_PROCESSING_IN_GENOMICS.pdf

12:00 → 13:00 Lunch break

13:00 → 15:00 Data processing in genomics, hands-on

Speaker: Dr Arnaud Lavergne (GIGA)

2021_GenomicsDataProcessing_handson.pdf

15:00 → 15:30 Coffee break

15:30 → 17:30 Matlab & Octave practical

Speakers: Prof. Christophe Phillips (GIGA in silico medicine & CRC in vivo imaging) , Dr Mohamed Bahri