Final Master Project

Master in Bioinformatics and Computational Biology, Universidad Autónoma de Madrid (UAM)
Barcelona Biomedical Genomics Lab (BBGLab) https://bbglab.irbbarcelona.org/
Institut de Recerca Biomèdica de Barcelona (IRB Barcelona)

Necessary code to reproduce all the data from the final master project entitled as:
"The impact and function of degron disruptions in c-terminal truncated proteins and its role in tumorigenesis"

Student:        Raquel Blanco Martínez-Illescas
Directors:      Mònica Sánchez Guixé and Núria López-Bigas
Academic tutor: Luis del Peso Ovalle
Course:         2021/2022

The following diagram indicates the workflow to perform the analysis, including data collection and preprocessing (yellow), downstream analysis (green) and data analysis (red). All the necessary code is stored in this repository, except the scripts used to generated CPTAC and CCLE preprocessed tables.

Content

• Repository structure
• Requirements
• Data collection and preprocessing
• Downstream analysis
• Data analysis

Repository structure

The repository has the following directories:

• data: contains any necessary data that cannot be generated with the scripts in scripts folder, except CPTAC and CCLE preprocessed tables. Also, contains manually edited data from external sources to adjust format to input requirements of some functions. Manually curated alignments for ELM-Manual PWMs generation are stored in elm_manual/alignments/curated in FASTA format and PNGs in elm_manual/alignments/curated_pngs.
• scripts: contains all the scripts needed to reproduce the analysis. Divided in:
  • data_collection_prepro: scripts to run data collection and preprocessing (yellow in the workflow diagram).
  • downstream_analysis: scripts to run downstream analysis, including generation of Position Weight Matrices (PWMs), de novo degron identification and mutation annotation (green in the workflow diagram).
  • data_analysis: scripts to run data analysis (red in the workflow diagram).
  • external: scripts from external sources.
  • Utils: general util functions for the analysis.

Requirements

All the analysis was performed using conda in an environment with the dependencies in environment.yml. To install the same environment execute:

conda env create -n degrons -f environment.yml

Data collection and preprocessing

1. UbiNet

1.1. Position Probability Matrices (PPMs)

ubinet_PWMs.ipynb

The first part of this Jupyter notebook contains the code to parse the HTML of UbiNet 2.0 database and retrieve the PPMs of each degron motif. This notebook is stored in scripts/downstream_analysis.

1.2. UbiNet presumed degrons

ubinet_degrons.ipynb

Jupyter notebook with the code to collect the presumed degrons used by UbiNet to generate the PPMs motifs (used only in the last part of the UbiNet PWMs validation).

2. ELM-Manual database

2.1. Database generation

create_elm_manual_database.ipynb

Jupyter notebook with the code to generate the ELM-Manual database of experimentally validated degrons.

2.2. Database preprocessing

prepro_elm_manual_database.ipynb

Jupyter notebook with the code to preprocess ELM-Manual database and generate the true degron sets.

3. Human proteome of Ensembl 92 canonical transcripts

ensembl_proteome.ipynb

Jupyter notebook with the code to preprocess the downloaded human proteome of Ensembl 92 canonical transcripts.

4. Last exons of Ensembl 92 canonical transcripts

ensembl_last_exons.ipynb

Jupyter notebook with the code to preprocess the downloaded exons of Ensembl 92 canonical transcripts and extract every gene's last exon.

5. Annotate Ensembl transcript stable ID (ENST) in CCLE and CPTAC datasets

stabch_annotate_enst.py

Python script to annotate every mutation or WT form with the ENST of the canonical transcript.

6. Annotate cancer type in CCLE dataset

add_ctype_to_CCLE.ipynb

Jupyter notebook with the code to annotate the cancer type to each sample.

Downstream analysis

1. Position Weight Matrices (PWMs) degron motifs

1.1. PWMs from UbiNet degron motifs

ubinet_PWMs.ipynb

The second part of this Jupyter notebook contains the code to transform the PPMs into PWMs.

1.2. PWMs from ELM-Manual degron motifs

elm_manual_PWMs.ipynb

Jupyter notebook with the code to align degron sequences per motif and transform curated alignments into PWMs. Also generates intermediate PPMs.

2. PWM scan

2.1. Scan

motifs_scan_proteome.py

Python script to scan a set of proteins (e.g.: proteome) with a PWM using a sliding window technique.

2.2. True degrons scan

motifs_separate_substrates.py

Python script to divide the motifs_scan_proteome.py output in true degrons and the rest of proteins.

3. UbiNet PWMs validation

ubinet_PWMs_validation.ipynb

Jupyter notebook with the code to evaluate UbiNet PWMs with ELM-Manual degrons.

4. PWM quality analysis (positivity threshold, information content, specificity and discovery activity)

motifs_quality_analysis.py

Python script to calculate per motif features and quality metrics.

5. PWM iterative enrichment

motifs_iterative_enrichment_degener.py

Python script to enrich ELM-Manual PWMs using E3 ligase-substrate interactions from UbiNet database.

6. De novo degron identification

6.1. Discovered degrons

motifs_discovered_degrons.py

Python script to extract the discovered degrons from the motifs_scan_proteome.py output.

6.2. Pool overlapping degrons

pool_overlapping_degrons.ipynb

Jupyter notebook with the code to define the overlapping discovered degrons and pool them together.

7. Mutations in the last exon

stabch_annotate_lastexon.py

Python script to annotate every last-exon mutation in CCLE and CPTAC datasets.

8. Mutation annotation in the discovered degrons

stabch_annotate_degrons.py

Python script to annotate CPTAC and CCLE mutations and WT forms in the discovered degrons. This script was executed in paralell for all CPTAC cancer types and CCLE using qmap with template file stabch_annotate_new_instances.qmap.

stabch_create_pancancer.py

Python script to merge CPTAC cancer types tables into a pancancer table after degron annotation.

Data analysis

Figures.ipynb

Jupyter notebook with the code to reproduce all the plots and statistical analysis in the figures of the manuscript.