Summer school: “Big data Analytics and Machine Learning Meets Omics”

In summer 2018 the faculties of Health and Life Sciences, and Science and Engineering, hosted a ten week summer school for students interested in data science, analytics and machine learning. The five successful students received supervision from leading experts in the field, as well as a stipend.

The projects focussed on the application of data science techniques in the area of biological and biomedical research, called “omics”. These areas include the study of genomes and related technologies that produce large amounts of complex data, suitable for data mining. Researchers at Liverpool are using these techniques for “personalised health” research (e.g. to understand how your genetics relate to your likelihood of getting a disease), for cancer research (e.g. for understanding how certain genes are switched on or off during cancer progression), and to understand how infectious diseases can spread or how we can better detect them.

“Insightful, collaborative, applied”

We asked the students to explain their research project and experience of the summer school:

Di LiuDi Liu (BSC Computer Science) worked on the project “Topological data analysis for bioinformatics”.

 Can you tell us a little bit about your current research project

The intention of the project is to visualize the gene-gene interactions by analysing correlation coefficient of gene pairs as the pair’s weight score.

To achieve this we are analysing glioma gene scores by MPI and visualizing the analysis’ results on Cytoscape.

 How have you found the project? What has been the highlight?

The project has been excellent! Working with Drs Kurlin and Krishna broadened my vision and enriched my experience of using the knowledge I have learned.

Also, the project group is awesome. Two supervisors and one teammate, they helped me to overcome any obstacles during my internship.

The highlight for me was successfully using the C++ knowledge to fix the problem we met in the project, and learning how to use MPI to complete parallel programming to save a lot computer running time.

Has anything surprised you about the project?

I was expecting a more reserved role and have solo solved the data part of tasks because of the trust shown in me by Drs Kurlin and Krishna.

Have you seen the University in a different light as a researcher, compared to being here as a student?

Yes, the research side of the University is awesome. On the one hand, it practised the knowledge I have learned at University. On the other hand, it gave me a chance to learn more skills that can’t be learned in lectures. I think this will assist me in the future studies.

Has the research project changed your expectations of the future?

This project is relational with my expectation of the future. So, it let me realize my future’s expectation more clearly. Besides, it also let me see more possibilities to connect my subject with other subjects.

If you were to describe the experience in 3 words, what words would you choose?

Enjoyable, teamwork, rewarding.

Weiyi Ren (BSC Mathematics) worked on the project “Building a statistical model to help quantify molecules important for cell signalling and cancer”

Can you tell us a little bit about your current research project?

The project aims to understand how genetics relate to the likelihood of getting a disease, and to understand how infectious diseases can spread or how we can better detect them. My task was to build a statistic model to predict the change of intensity or retention time between the modified or unmodified peptide.

How have you found the project? What has been the highlight?

The team lead by Professor Andy Jones was searching for undergraduate students to help them build a statistic model. We tried several new approaches: random forest and neural network. The results were very successful comparing with the linear regression model.

Has anything surprised you about the project?

The project was very smooth. We met each week to talk about the progress. The supervisors were very passionate to help us with our difficulties. We could get detailed explanations from them on time.

Have you seen the University in a different light as a researcher, compared to being here as a student?

Yes, the research side of the University is excellent. Participating in a research is very different from being a student. It involves much more communication between the supervisor and the students, which would be no doubt a valuable experience for one’s future academic life.

Has the research project changed your expectations of the future?

Yes, this experience provides me with more possibilities for my future career. Data science is definitely a very good choice for me.

If you were to describe the experience in 3 words, what words would you choose?

Enjoyable, productive, full.

Ash Myall photo.jpg BSC e-Finance student Ash Myall worked on the project “Distributed computing and analytics to annotate the human genome”.

Can you tell us a little bit about your current research project

The intention of the project has been to develop a web-based visualization suite for results of Peptide Search Engines within proteomics analysis, focused specifically on the crowdsourced search engine which operates on a distributed network.

To develop this, we have used R and shiny, which together can quickly produce interactive dynamic visualizations.

In lay terms….producing visual representations of large scale data sets, to identify trends in the behaviour of proteins in samples (e.g. the human body).

How have you found the project? What has been the highlight?

The project has been incredibly insightful, which subsequently lead me to pursue postgraduate research in the field. My favourite part has been consolidating my understanding and confidence in using R.

Has anything surprised you about the project?

Yes, before I had little idea how essential data visualisations can be to understanding underlying trends, and how useful making interactive plots is; questions can be answered instantly about the data by a user.

Have you seen the University in a different light as a researcher, compared to being at here as a student?

Yes, I’ve found the independence to give greater flexibility in finding solutions to unseen problems.

Has the research project changed your expectations of the future?

Absolutely, I’m about to start an MRes in Advanced Biological Sciences with the bio informatics pathway following my interest in biological data analysis on the project.

If you were to describe the experience in 3 words, what words would you choose?

Insightful, collaborative, applied.

Is there anything else you would like add?

This has been a great opportunity, I’d recommend it to anyone who’s got an interest in biological sciences and comes from a computational background like myself. Not only has it provided an excellent application from my undergraduate skills, but it’s also set me in a great position for beginning a career or for going into a PhD.

Antoine Rohmer.jpgAntoine Rohmer (BSC Computer Science) worked on the project “Exploiting Graph Databases for bioinformatics” which included an internship at IBM.

Can you tell us a little bit about your current research project

The purpose of the project was to gain an understanding of the technologies underlying graph databases and parallel computing to create a large-scale visualization of gene-gene interactions from data provided by the University.

The technologies used in the project include Neo4j and the Cypher querying language on the graph database side. C++ and Python were used in tandem with supercomputers provided by IBM for data processing. Cytoscape was used for visualization. The open source pathway database reactome was used as the foundation of the database created for the project.

 How have you found the project? What has been the highlight?

The project was a great foray into the professional world. Working alongside experts in the field and absorbing their insight was a fantastic learning experience.

The highlight of the project for me was the moment the graph database and parallel computing sides of the project combined to form a coherent whole that until then seemed intangible.

 Has anything surprised you about the project?

Perhaps the biggest surprise was how largely independent I was in pursuing the project. Although a plan and structure existed, most of the meeting setups, deadlines, and choices in technologies were up to me. This in particular was a refreshing excursion from the otherwise rigid structure found in my studies.

 Have you seen the University in a different light as a researcher, compared to being here as a student?

Although at times the project felt like a continuation of my studies due to the large amount of research that had to be conducted, the knowledge that the end product will be used as a stepping stone to other research is a much greater motivator than grades.

 Has the research project changed your expectations of the future?

The project truly validated my interests in data analysis and got rid of any doubts previously had about which postgraduate program to pursue. More importantly it made me even more eager to enter the professional world.

If you were to describe the experience in 3 words, what words would you choose?

Insightful, refreshing, validating

Advertisements

Student’s Biochemical Society summer placement in IIB

pic1

Christopher Steel exploring the role of proteoglycans and glycosaminoglycans in aortic aneurysm and dissection funded by Biochemical Society supervised by Hannah Davies and Jill Madine

Aortic aneurysm is a bulge in a section of the aorta. This causes weakening of the aortic wall and increased risk of the blood vessel bursting. A number of biochemical factors can complicate the local environment of an aorta which then has a direct consequence on the mechanical function of the vessel. One factor reported to affect the integrity of the vessel wall is the concentration and distribution of proteoglycans (PGs) and glysaminglycans (GAGs), particularly ‘pooling of PGs’ which may compromise the aorta wall and lead to delamination, aortic aneurysm, and dissection.

In his project Chris investigated the levels of PGs and GAGs in human tissue obtained during aortic surgery at Liverpool Heart and Chest Hospital using a range of gel-based techniques and dot-blots and employed immuno-staining to investigate the distribution of GAGs within the medial layer of the aorta wall. Data is being correlated with existing data on biomechanical properties and biochemical composition of collagen and elastin, along with clinical characteristics for the same patients to assess whether PG and GAG distribution could be contributing to altered aortic wall integrity in disease. PGs and GAGs could provide a future therapeutic biomarker to predict risk of aortic disease and rupture.

 

Wellcome Summer Studentship Award

Megan Souness completed a Wellcome Trust funded project with Elliott Stollar this summer titled “The role of electrostatic interactions in SH3 domain structure and function”. Megan was successful in purifying and thermodynamically characterising over 20 SH3 domains. She was also able to study the role salt played on the folding kinetics of the SH3 domain from Abp1p and found that salt affects the folding rate much more than the unfolding rate. Her work will be included in a publication which is hoped to be submitted soon.

Summer in the Dart Lab

Jagjit Binning (now Year 3 Biological Sci) and Mohamed Omar (now Year 3 Biochemistry) joined Caroline Dart’s group for 3 weeks in July to look at the expression and function of a specialized calcium influx pathway (the Orai/STIM system) in human skin cell lines.  This involved aseptic cell culture, SDS PAGE /immunoblot analysis and also learning how to record the tiny electrical currents carried by the movement of calcium ions using patch clamp electrophysiology.  The electrophysiological recordings are technically difficult and time-consuming so well done to Jagjit and Mohamed both managing to seal a microelectrode onto the surface of an individual cell at virtually their first attempt!

Summer in the Betancourt Lab with Nuffield and undergraduate students

Summer in the Betancourt Lab with Nuffield and undergraduate students

The Betancourt Lab hosted several students this summer, including two Nuffield students.  These were Emily Clarke and Mecmillan “Mec” Rajaratnam, who used PCR to survey Wolbachia in flies.

Emily said about the placement:  “I feel that I have benefitted greatly from my Nuffield placement and I have enjoyed every minute. I have gained a whole new platform of scientific skills, knowledge and techniques as well as been given an insight into the work that goes on within research science. Before beginning the placement, I did not really know what research science consisted of on a day to day basis however, experiencing this first hand has educated me on the works of scientific research daily. The project that I completed I thought was really interesting and I looked forward to attending every day of the placement to get some more results; in order to be one step closer to completing the project. Overall, I am so glad I was offered this fantastic opportunity and I definitely feel like I have gained so much from the experience.”

The Betancourt Lab also played host to second year undergraduate student Harry Collier, who carried out a project looking at the phenotypic effects of transposable element insertions in Drosophila. Harry also learned to code in Python during the course of his project.

XJTLU student Xiaoyue Song’s summer in IIB

XJTLU student Xiaoyue Song’s summer in IIB

This summer IIB’s Dr Dan Rigden hosted XJTLU student Xiaoyue Song:

Gut bacteria composition and activity can have big effects on human health and disease. There is therefore a lot of interest in understanding proteins and protein families encoded by the principal gut bacteria. Structural genomics has contributing by solving many structures but some of these have been deposited in databases without associated publications fully describing or predicting their functions. Xiaoyou has been focusing on a single family commonly found in gut bacteria, using a range of bioinformatics methods to try to predict its function. She strongly suspects that it is involved in carbohydrate metabolism and is currently trying to pin down the function further.

 

 

Are sleep patterns affected by the battle of the sexes? A summer project in IIB

Are sleep patterns affected by the battle of the sexes?  A summer project in IIB

Guest post by Elysse Hendrick, undergraduate summer student

Working with supervisor Tom Price and his team, I carried out a research project investigating whether sleep patterns are affected by the battle of the sexes. The idea is that in species where females mate with multiple males, it is in the male’s interest if any female he mates with lays many eggs as quickly as possible, because after a while she will mate with another male who will father most subsequent eggs. So if males could force females to be active all day laying eggs, this would benefit the male. Normally, fruit flies are active only at dawn and dusk, so daylight egg laying might mean a female risks death due to predators or drying out. To test this, I put mated and unmated fruit flies into an activity monitor which, using an infrared beam, monitored the level of activity of each fly over a week. Entering this project, I had extremely limited experience of working with flies, so this project allowed the development of many new skills, including: sexing and identifying flies using microscopy, performing mating assays, food-making and maintaining fly stocks. Additionally, undertaking this project helped in gaining confidence, both in working in a different laboratory setting than usual with unfamiliar equipment (such as the activity monitor), and in working independently. Aside from my project, I also helped and worked alongside Tom Price’s team and attended team meetings, which helped me gain a valuable insight into the dynamic of a research group and the roles everyone plays. I enjoyed my project and working with the team very much and would definitely recommend taking part in a summer project.