/GDA

Genome Data Analytics (DSSC 2021)

Primary LanguageHTML

Genomic Data Analytics (GDA)

Giulio Caravagna (gcaravagna@units.it) 3/5/2021

MSc program in Data Science and Scientific Computing. University of Trieste, Italy

Lecturers

Invited guest lecturers

  • Dr Riccardo Bergamin, University of Trieste
  • Dr Alex Graudenzi, CNR.
  • Dr Salvatore Milite, University of Trieste
  • Dr Daniele Ramazzotti, University of Milan-Bicocca.

Program

Lecturer Title When
Caravagna Variant calling from bulk sequencing 9/4
Caravagna Measuring aneuploidy from bulk sequencing 12/4
Caravagna Integrated quality control of somatic calls 16/4
Bergamin Population genetics for cancer 19/4
Caravagna Tumour subclonal deconvolution 21/4
Ramazzotti Somatic mutational signatures 23/4
Bergamin, Milite Basics of Single-cell RNA analysis 30/4
Graudenzi Longitudinal evolution from single cell 3/5
Milite Count-based models for single-cell data 5/5
Caravagna Evolutionary based stratifications 12/5
Caravagna Population-level models 14/5

Preamble

Part 1 - Somatic calling from bulk sequencing

Lecture: Variant calling from bulk sequencing (https://www.dropbox.com/s/2ngbbxiudux8h9v/Somatic_calling_annotated_lecture.pdf?dl=0)

  • (Theory) Mutation calling:

    • Tumour matched-normal design,
    • High-level design of GATK
    • Joint calling model

  • (Practice) Example VCF and PCAWG:

    • VCF manipulation
    • 27 PCAWG cases (mutation types, burden, etc.)

  • Readings (https://www.dropbox.com/s/bx2kam7tlf7tl5x/readings.zip?dl=0)

    • (tool) Roth, Andrew, et al. “JointSNVMix: a probabilistic model for accurate detection of somatic mutations in normal/tumour paired next-generation sequencing data.” Bioinformatics 28.7 (2012): 907-913.
    • (tool) Kim, Sangtae, et al. “Strelka2: fast and accurate calling of germline and somatic variants.” Nature methods 15.8 (2018): 591-594.
    • (tool) Benjamin, David, et al. “Calling somatic snvs and indels with mutect2.” BioRxiv (2019): 861054.
    • (tool) Rimmer, Andy, et al. “Integrating mapping-, assembly-and haplotype-based approaches for calling variants in clinical sequencing applications.” Nature genetics 46.8 (2014): 912-918.
    • (tool) GATK (Broad Institute)
    • Training: www.csc.fi/en/web/training/-/gatk2019
    • Lectures: https://www.youtube.com/watch?v=sM9cQPWwvn4&list=PLjiXAZO27elDHGlQwfd06r7coiFtpPkvI

Lecture: Measuring aneuploidy from bulk sequencing (https://www.dropbox.com/s/nberoeiisgmwknl/2.CNA_lecture.pdf?dl=0)

  • (Theory) Aneuploidy and Copy Number calling:

    • Motivation
    • ASCAT model
    • Segmentation

  • (Practice) Example runs with different tools:

    • ASCAT
    • Sequenza (inspection of alternative solutions)
    • Circular Binary Segmentation
    • Cohort-level distribution of CNAs per chromosome (length, percentage, copy state).

  • Readings (https://www.dropbox.com/s/ohz5f7e51dwpg71/readings.zip?dl=0)

    • (tool) Favero, Francesco, et al. “Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data.” Annals of Oncology 26.1 (2015): 64-70.
    • (tool) Van Loo, Peter, et al. “Allele-specific copy number analysis of tumors.” PNAS 107.39 (2010): 16910-1691
    • (tool) Ross, Edith M., et al. “Allele-specific multi-sample copy number segmentation in ASCAT.” Bioinformatics (2020).
    • (tool) Olshen, Adam B., et al. “Circular binary segmentation for the analysis of array‐based DNA copy number data.” Biostatistics 5.4 (2004): 557-572.
    • (Review) Ben-David, Uri, and Angelika Amon. “Context is everything: aneuploidy in cancer.” Nature Reviews Genetics 21.1 (2020): 44-62
    • (Review) Weaver, Beth AA, and Don W. Cleveland. “Does aneuploidy cause cancer?.” Current opinion in cell biology 18.6 (2006): 658-667.
    • (In vivo measurements) Bolhaqueiro, Ana CF, et al. “Ongoing chromosomal instability and karyotype evolution in human colorectal cancer organoids.” Nature Genetics 51.5 (2019): 824-834.
    • (coding) DNAcopy: A Package for Analyzing DNA Copy Data https://bioconductor.org/packages/release/bioc/vignettes/DNAcopy/inst/doc/DNAcopy.pdf
    • (coding)Total copy-number segmentation using CBS. https://cran.r-project.org/web/packages/PSCBS/vignettes/CBS.pdf

Lecture: Integrated quality control of somatic calls

Parte 2 - modelling and inference from bulk

Lecture (R Bergamin): Population genetics models of growth

  • (Theory) Branching processes and other models (https://www.dropbox.com/s/wj7qdlg3yrdwwp7/branchig_process.pdf?dl=0)

    • Cancer Evolution as Stochastic Branching Process
    • Markov System and Master equation
    • Some Examples: Moran Model, Wright-Fisher Model, Coalescence
    • Birth-Death Process
    • Luria-Delbruck Model
    • Theory of 1/f tail
    • Quantify Cancer Evolution from VAF Spectrum
    • Spatial Tumor Growth

  • (Practice) Tumour growth simulation:

  • Readings (https://www.dropbox.com/s/ydl8zl0rhd46lx3/references_branching_process.zip?dl=0)

    • Turajlic, et al., "Resolving genetic heterogeneity in cancer", Nature Reviews Genetic Volume Issue 2019
    • Beerenwinkel, et al., “Cancer Evolution: Mathematical Models and Computational Inference”, DOI:10.1093/sysbio/syu081,
    • Kessler and Levine, “Large Population solution of the stochastic Luria-Delbruck evolution model ”, PNAS Volume 110 issue 29 2013
    • Weber and Frey, “Master equations and the theory of stochastic path integrals”, arXiv:1609.02849v2, 2 April 2017
    • Durret, “Branching Process Models of Cancer”, Mathematical Biosciences Institute Lecture Series 1.1 Stochastics in Biological Systems
    • Willimas, et al., “Identification of neutral evolution accross cancer type”, Nature Genetics, 2016.
    • Williams, et al., “Quantification of subclonal selection in cancer from bulk sequencing data ”, Nature Genetics, 2019.

Lecture: Tumour subclonal deconvolution

  • (Theory) Subclonal deconvolution (https://www.dropbox.com/s/l7efzgiv6o8jy3p/5.MOBSTER.pdf?dl=0):

    • Tail modelling versus subclones
    • Read counts analysis
    • Multi-sample deconvolution

  • (Practice) Deconvolution in practice

    • MOBSTER runs with WGS data

  • Readings (https://www.dropbox.com/s/ijn3edftzys8bcy/readings.zip?dl=0)

    • Same as previous lecture
    • Roth, Andrew, et al. “PyClone: statistical inference of clonal population structure in cancer.” Nature methods 11.4 (2014): 396-398.
    • Caravagna, et al., "Subclonal reconstruction of tumors by using machine learning and population genetics", Nature Genetics 52, 2020.

Lecture (D Ramazzotti): Mutational signatures in human cancers

  • Theory: (https://www.dropbox.com/s/t60cjdipp93cueq/6.%20Signatures.pdf?dl=0)

    • Concepts behind mutational signatures
    • De novo inference of mutational signatures
    • Solving with non-negative matrix factorization (NMF)
    • Mutational signature extraction from pan-cancer data

  • Practice (install required packages before the lecture):

    • Examples and best practice on real data
    • Analysis of breast cancer data

Readings:

  • (Concepts on mutational signatures) Alexandrov, Ludmil B., et al. "Signatures of mutational processes in human cancer." Nature 500.7463 (2013): 415-421.
  • (Concepts on mutational signatures) Alexandrov, Ludmil B., et al. "The repertoire of mutational signatures in human cancer." Nature 578.7793 (2020): 94-101.
  • (Tool - SigProfiler) Alexandrov, Ludmil B., et al. "Deciphering signatures of mutational processes operative in human cancer." Cell reports 3.1 (2013): 246-259.
  • (Tool - SparseSignatures) Lal, A., et al. "De Novo Mutational Signature Discovery in Tumor Genomes using SparseSignatures." (2020).
  • (Statistics - Non-negative matrix factorization) Brunet, Jean-Philippe, et al. "Metagenes and molecular pattern discovery using matrix factorization." Proceedings of the national academy of sciences 101.12 (2004): 4164-4169.
  • (Statistics - Non-negative matrix factorization) Owen, Art B., and Patrick O. Perry. "Bi-cross-validation of the SVD and the nonnegative matrix factorization." The annals of applied statistics 3.2 (2009): 564-594.

Part 3 - Single-cell sequencing

Lecture (R Bergamin, S Milite): Basics of Single-cell RNA analysis

Lecture

Lecture (A Graudenzi, F Angaroni and D Maspero): Longitudinal evolution from single cell

  • Theory: Inference of phylogenies from single cell data (https://www.dropbox.com/s/isn7qgz1l661gax/Single-cell-phylo.pptx?dl=0)

    • Perfect phylogenies from categorical data: the Gusfield algorithm,
    • Translating the perfect phylogeny problem as non-negative factorization (NMF)
    • Technical noise (sequencing errors) and biological variability: the need for probabilistic models of clonal evolution.
    • The likelihood function and the probabilistic graphical model of SCITE
    • Estimation of the error rate
    • Structure learning via MCMCd
    • Extension: longitudinal models (LACE)
    • Extension: modeling mutation losses (SIFIT)
    • Extension: including population dynamics (SICLONEFIT)

  • Practice (Data to download https://www.dropbox.com/s/sy8gadzjex8luhk/Melanoma.zip?dl=0):

  • Readings

    • [Gusfield] “Efficient algorithms for inferring evolutionary trees” D. Gusifield 1991
    • [SCITE] ”Tree inference for single-cell data” Janh et al. 2016
    • [LACE] “Longitudinal cancer evolution from single cell” D. Ramazzotti et al. 2020
    • [SIFIT] ”SiFit: inferring tumor trees from single-cell sequencing data under finite-sites models” H. Zafar et al. 2017
    • [SICLONEFIT] ”SiCloneFit: Bayesian inference of population structure, genotype,and phylogeny of tumor clones from single-cell genome sequencing data” H. Zafaret al. 2019

Lecture (S Milite): Count-based models for single-cell data

  • Theory (https://www.dropbox.com/s/6p8le8wesf2o6np/scRNA_seq_lesson.pdf?dl=0):

    • Generative modelling as an alternative to pipelines
    • Poisson and Negative binomial distributions
    • Count based modelling, RNA-seq vs scRNA-seq
    • Count models for normalisation (scTransform)
    • Scaling NB models with variational autoencoders (scVI)
    • CONGAS (genotype CNV from scRNA-seq)
    • Elements of Gradient based variational inference
    • Discrete Latent Variable modelling

  • Practice:

    • Example run of CONGAS on breast cancer 10x dataset

  • Readings:

    • Hafemeister, Christoph, and Rahul Satija. ‘Normalization and Variance Stabilization of Single-Cell RNA-Seq Data Using Regularized Negative Binomial Regression’. Genome Biology 20, no. 1 (23 December 2019): 296. https://doi.org/10.1186/s13059-019-1874-1.
    • Jang, Eric, Shixiang Gu, and Ben Poole. ‘Categorical Reparameterization with Gumbel-Softmax’. ArXiv:1611.01144 [Cs, Stat], 5 August 2017. http://arxiv.org/abs/1611.01144.
    • Kingma, Diederik P., and Max Welling. ‘Auto-Encoding Variational Bayes’. ArXiv:1312.6114 [Cs, Stat], 1 May 2014. http://arxiv.org/abs/1312.6114.
    • Lopez, Romain, Jeffrey Regier, Michael B. Cole, Michael I. Jordan, and Nir Yosef. ‘Deep Generative Modeling for Single-Cell Transcriptomics’. Nature Methods 15, no. 12 (December 2018): 1053–58. https://doi.org/10.1038/s41592-018-0229-2.
    • Milite, Salvatore, Riccardo Bergamin, and Giulio Caravagna. ‘Genotyping Copy Number Alterations from Single-Cell RNA Sequencing’. BioRxiv, 1 January 2021, 2021.02.02.429335. https://doi.org/10.1101/2021.02.02.429335.
    • Sarkar, Abhishek, and Matthew Stephens. ‘Separating Measurement and Expression Models Clarifies Confusion in Single Cell RNA-Seq Analysis’. BioRxiv, 1 January 2020, 2020.04.07.030007. https://doi.org/10.1101/2020.04.07.030007.
    • Schulman, John, Nicolas Heess, Theophane Weber, and Pieter Abbeel. ‘Gradient Estimation Using Stochastic Computation Graphs’. ArXiv:1506.05254 [Cs], 5 January 2016. http://arxiv.org/abs/1506.05254.
    • Svensson, Valentine. ‘Droplet ScRNA-Seq Is Not Zero-Inflated’. Nature Biotechnology 38, no. 2 (February 2020): 147–50. https://doi.org/10.1038/s41587-019-0379-5.

Part 4 - Population-level inference

Lecture: Evolutionary based stratifications

  • (Theory) Detecting repeated evolution from multi-region bulk sequencing

    • Clone-trees and tree expansion
    • Expectation Maximisation for latent model discovery
    • Evolutionary distance and cluster

  • (Practice) Inference in practice

    • Colorectal adenomas with REVOLVER
    • TRACERx Lung Adencarcinomas with REVOLVER

  • Readings

    • Caravagna, Giulio, et al. “Detecting repeated cancer evolution from multi-region tumor sequencing data.” Nature methods 15.9 (2018): 707-714.

Lecture: Population-level models

  • (Theory) Bayesian Networks models

    • Conjunctive Bayesian Networks
    • Suppes’ probabilistic causation

  • (Practice) Inference in practice

    • Analysis of CODREAD with PICNIC
    • Analysis of other cbio data

  • Readings

    • Beerenwinkel, Niko, Nicholas Eriksson, and Bernd Sturmfels. “Conjunctive bayesian networks.” Bernoulli (2007): 893-909.
    • Gerstung, Moritz, et al. “Quantifying cancer progression with conjunctive Bayesian networks.” Bioinformatics 25.21 (2009): 2809-2815.
    • Caravagna, Giulio, et al. “Algorithmic methods to infer the evolutionary trajectories in cancer progression.” Proceedings of the National Academy of Sciences 113.28 (2016): E4025-E4034.
    • Ramazzotti, Daniele, et al. “CAPRI: efficient inference of cancer progression models from cross-sectional data.” Bioinformatics 31.18 (2015): 3016-3026.
    • Loohuis, Loes Olde, et al. “Inferring tree causal models of cancer progression with probability raising.” PloS one 9.10 (2014): e108358.