/01-genetic-interactions-in-cancer

Project proposal

Project 01: Genetic interactions and cancer cell survival

Data Sciece project based teaching - 2019

Project overview and guidelines

Supervisor:

Tutor:

Introduction

Genes rarely act alone, but rather in concert with others. In a signalling or a metabolic pathway, a single gene (and its translated product - protein) is not reflective of the final signalling or metabolic pathway activity. However, different genes act together (some activating, some repressing while others neutral) resulting in the net pathway output.

The aim of this project is to identify genetic interactions and their consequent effect on cancer cell survival. Cancer is a disease of accumulated defects (mutations, copy number changes, aneuploidies etc) in the genome. While the number of genomic defects in different cancers might vary from very few to thousands, only a handful of them are driving the process of oncogenesis. These driver mutations are observed in the highest frequencies, significantly enriched over the background mutation rates while the rest are typically irrelevant passenger mutations.

In practice, these driver genes can mostly not be targeted by drugs. This is mainly because the driver genes are often highly essential for normal cell survival too. As a result, the anti-cancer drugs cannot discriminate between cancer and normal cells. An alternative strategy consists in targeting some other genes that interact with the driver genes in the cancer cell but not in the normal cells.

Fig. 1 - Genetic interactions in cancer

Fig. 1 - Genetic interactions in cancer

In Fig.1, some of the possible genetic interactions occurring in a cancer cell has been summarized. Lets imagine a driver mutation (red) that transforms a normal cell into a malignant cell which divides uncontrollably. However, its wild type (not mutated) counterpart (black) divides normally. Now, we cannot target the driver mutation (red) because it is also essential in the normal cells. So we instead try to find another gene (yellow, green and blue) which, when perturbed in combination with the driver mutation, reduces the cancer cell proliferation while having no effect in the wild type condition upon the same perturbation.

As a basis for this project, we will use a large dataset of cancer cell lines including a combination of gene expression data, information about mutations and copy number changes, as well as data on the sensitivity of the cell lines when certain genes are knocked-down (see below the detailled description).

Objective

The objective of this project is to achieve the following:

  • Choose a cancer type of interest (Fig. 2) with a complete genomic profile i.e. a cancer type with gene expression, gene copy number, gene mutation and CRISPR/Cas9 based gene knockdown data. The cancer type of interest must have at lest 20 representative cell lines.

  • For the chosen cancer type, review recent literature to identify 3-5 most prominent mutations or genomic alterations driving the cancer growth (for example mutations in p53, EGFR, RB etc). A good starting point might be the publications from the TCGA consortium

  • Ideally, chose those driver alterations which are difficult to target clinically (for example p53)

  • Identify appropriate cell line models to study the specific genomic alteration in a cancer type of interest. For instance, if you chose to focus on p53 mutation for a particular cancer type, segregate cell lines for that cancer type with or without p53 mutations. This way you can check many of your hypothesis like

    • What happens when there is p53 mutation vs no p53 mutation
    • If I observe something in p53 mutated scenario, do I still observe this in a non p53 mutated scenario
    • How confident am I to claim that something is exclusively p53 mutation related effect etc

  • In the cancer type of your interest, for the chosen driver alteration, find which are the most correlated mutations or copy number changes associated with the driver alterations.

  • Find if there is functional/biological relationship between these correlated alterations. For example are they in the same pathways as the driver mutation, are they part of the same protein complex etc. Use relevant publicly available databases to get these information.

  • Find if there are clinically approved drugs for any of the correlated alterations. Use relevant publicly available databases to get these information.

  • Make a coherent report of your complete analysis and results using R markdown.

Fig. 2 - Number of cell lines per cancer type

Fig. 2 - Number of cell lines per cancer type

Description of datasets

We have provided a R list object (as a .RDS file). You can find the data using this link: https://figshare.com/s/caa3448439fcf6032df4

Once you have downloaded it, load it as below:

allDepMapData = readRDS("path/to/your/directory/DepMap19Q1_allData.RDS")
names(allDepMapData)
[1] "expression" "copynumber" "mutation"   "kd.ceres"   "kd.prob"    "annotation"

The list named as allDepMapData consists of the following matrices -

  • Gene expression
    • This matrix consist of gene TPM (transcripts per million) values. These values reflect the level of gene expression, higher values suggests over expression of genes and vice versa. The rows of this matrix are the gene names and the columns are the cancer cell line identifiers.
  • Gene copy number
    • This matrix consist of gene copy number (CN) values. In absolute terms, CN = 2, since there are two alleles per genes. In cancer, genes might be amplified CN > 2 or deleted CN < 2. These values reflect the copy number level per gene, higher values means amplification of genes and vice versa. The rows of this matrix are the gene names and the columns are the cancer cell line identifiers.
  • Gene mutations
    • Is the annotation file for the various mutations observed in a sample. The isDeleterious flag specifies if mutation has a functional effect or not.
  • Gene knockdown (CERES)
    • This matrix consist of gene knockdown scores. The score is a measure of how essential/important is a particular gene for the cell survival. This score reflects whether upon knocking down that genes does the cell reduce its proliferation or increases it or has no change. Smaller values refers to higher essentiality. The rows of this matrix are the gene names and the columns are the cancer cell line identifiers.
  • Gene knockdown (Probability)
    • The probability value for the effect of the gene knockdown. A higher probability signifies that knocking down that particular gene very likely reduces cell proliferation.
  • Annotation
    • This matrix gives information regarding the cell lines. The DepMap_ID provides the uniform cell line identifiers used in all the data sets below. The CCLE_Name is encoded as cell line name _ Tissue of origin. The columns Primary.Disease and Subtype.Disease refers to the tissue/tumor of origin.

Literature review

Reviews

Read these reviews to gain some understanding of genetic interactions in cancer.

Experimental methods

An experimental example of killing ovarian cancer cells specifically with ARID1A mutation by inhibiting BRD2.

Computational methods

Read these papers to understand how we can computationally identify novel genetic interactions in cancer

How to structure your project

Project proposal

You first task wil we to define a project proposal, which should include

  • summary of literature on this dataset
  • questions you want to address
  • approximate timetable

You will present this project proposal together with a literature review on the subject 3 week after the begining of the semester (10 minute presentation + 5 minutes discussion).

Project

You project MUST contain the following elements:

  • descriptive statistics about the datasets
  • graphical representations
  • dimension reduction analysis (PCA, clustering or k-means)
  • statistical tests (t-test, proportion tests etc)
  • linear regression analysis, either uni- or multivariate

Data cleanup

You will be analyzing multiple genomic data sets together. It is essential that you explore each dataset and clean it. Cleaning can refer to many things:

  • Removing missing values
  • Imputing missing values
  • Removing low variance columns/rows
  • Removing batch effects
  • Removing outlier samples (only if it is due to technical issues !!)
  • Making sure that data is in the correct format, for example, numbers should be encoded as numeric and not as characters. Categorical variables should be factors etc.
  • Re-ordering rows/columns in meaningful and useful ways

Data exploration

Now that you have cleaned data, explore your data to understand its structure. Perform basic exploratory data analysis.

  • Look at the distribution of the overall data, specific samples or features.
  • Visualize the data distribution
  • Visualize the inter-dependencies among specific samples/features of interest
  • Check some of your hypothesis like - is something high/low between two conditions etc

Data reduction

You have a high dimension matrix, that is, you have way more features (~25000 genes) than observations (~30 samples).

  • Try out methods to reduce the dimensionality of this data.
  • Cluster your samples to identify similar and dis-similar groups
  • Check how well the groups separate based on the features of your interest

Data modelling

Imagine how you could try to predict sensitivity of cell lines to knockdown using other features as input. Test how well this might work. Alternatively, can you predict the impact of gene mutations or copy number alterations on the expression of that gene?

Final presentation

You will have to present your findings in a 15 min presentation followed by 15 min of questioning

Scoring

You will be graded on -

  • Quality of project proposal (realistic vs ambitious vs planning)
  • Teamwork and individual contributions
  • Completion of essential statistical tasks
  • Understanding of concepts
  • Troubleshooting
  • FInal presentation