Skip to content

Latest commit

 

History

History
executable file
·
836 lines (592 loc) · 131 KB

README.md

File metadata and controls

executable file
·
836 lines (592 loc) · 131 KB
Raw

A continually expanding collection of cancer genomics tools and data

MIT License PR's Welcome

Please, contribute and get in touch! See MDmisc notes for other programming and genomics-related notes.

Table of content

Drugs

  • Drug-target interaction (DTI) predictions using the Bidirectional Encoder Representations from Transformers (BERT) algorithm. 2.1M studies from PubMed (PubTator API). 512-token (word) sequences. Majority voting of five BERT modelts (BERT, SciBERT, BioBERT, BioMed-RoBERTa and BlueBERT). 99% accuracy using 10-fold CV. Other DTI resources: ChEMBL, BindingDB, PubChem, GtopDB, and DrugTargetCommons. Predicted and integrated data download, text format. Main portal: Drug Target Commons (DTC) - a crowd-sourcing platform to improve the consensus and use of drug-target interactions.
    Paper Aldahdooh, Jehad, Markus Vähä-Koskela, Jing Tang, and Ziaurrehman Tanoli. “Using BERT to Identify Drug-Target Interactions from Whole PubMed.” Preprint. Bioinformatics, September 11, 2021. https://doi.org/10.1101/2021.09.10.459845.
  • Cancer Perturbed Proteomics data - Protein responses to drug perturbations across cancer cell lines. Approx. 210 clinically relevant proteins. A systematic map of protein-drug connectivity. Download.
    Paper Zhao, Wei, Jun Li, Mei-Ju M. Chen, Yikai Luo, Zhenlin Ju, Nicole K. Nesser, Katie Johnson-Camacho, et al. “[Large-Scale Characterization of Drug Responses of Clinically Relevant Proteins in Cancer Cell Lines](https://doi.org/10.1016/j.ccell.2020.10.008).” Cancer Cell, (December 2020)
  • CARE - biomarker identification from interactions of drug target genes with other genes. Multivariate linear modeling with interaction term. Illustrative example of interaction of BRAF mutation and EGFR expression. Sample separation by gene expression correlation with CARE score better predicts survival. Comparison with correlation, elastic net, support vector regression. Download, nls_logsig tool to compute AUC for dose curves.
    Paper Jiang, Peng, Winston Lee, Xujuan Li, Carl Johnson, Jun S. Liu, Myles Brown, Jon Christopher Aster, and X. Shirley Liu. “[Genome-Scale Signatures of Gene Interaction from Compound Screens Predict Clinical Efficacy of Targeted Cancer Therapies](https://doi.org/10.1016/j.cels.2018.01.009).” Cell Systems 6, no. 3 (March 2018)
  • CancerRxGene - Drug-gene targets. Lots of drug sensitivity information.
    Paper Yang, Wanjuan, Jorge Soares, Patricia Greninger, Elena J. Edelman, Howard Lightfoot, Simon Forbes, Nidhi Bindal, et al. “[Genomics of Drug Sensitivity in Cancer (GDSC): A Resource for Therapeutic Biomarker Discovery in Cancer Cells](https://doi.org/10.1093/nar/gks1111).” Nucleic Acids Research 41, no. Database issue (January 2013)
  • CellMinerCDB - genomics (gene expression, mutations, copy number, methylation, and protein expression) and pharmacogenomics (drug responses and genomics interplay) analyses of cancer cell lines. Integrates NCI-60, GDSC, CCLE, CTRP, and NCI-SCLC databases built on top of rcellminer R package. Correlation and multivariate analyses. Tissue-specific analysis. 10m video tutorial
    Paper Rajapakse, Vinodh N., Augustin Luna, Mihoko Yamade, Lisa Loman, Sudhir Varma, Margot Sunshine, Francesco Iorio, et al. “[CellMinerCDB for Integrative Cross-Database Genomics and Pharmacogenomics Analyses of Cancer Cell Lines](https://doi.org/10.1016/j.isci.2018.11.029).” IScience 10 (December 2018)
  • CDGnet - targeted therapies recommendation system. Input - text file with molecular alterations. Integrating them with KEGG pathways, FDA-approved drugs, DailyMed, DrugBank, PubChem.Four drug prioritization categories: indication for the same cancer type, different type, pathway-guided, relevant in other cancer types. GitHub. Similar functionality - PreMedKB.
    Paper Kancherla, Jayaram, Shruti Rao, Krithika Bhuvaneshwar, Rebecca B. Riggins, Robert A. Beckman, Subha Madhavan, Héctor Corrada Bravo, and Simina M. Boca. “[An Evidence-Based Network Approach to Recommending Targeted Cancer Therapies](https://doi.org/10.1101/605261).” Bioinformatics, April 11, 2019.

  • CTRP - The Cancer Therapeutics Response Portal (CTRP) links genetic, lineage, and other cellular features of cancer cell lines to small-molecule sensitivity with the goal of accelerating discovery of patient-matched cancer therapeutics.

  • DepMap - Large-scale RNAi screen for cancer vulnerability genes in 501 cell lines from 20 cancers, shRNA silencing ~17,000 genes. DEMETER - Modeling and removal of shRNA off-target effects. 6 sigma cutoff of DEMETER scores to identify 769 differential gene dependencies. ATLANTIS model to predict other genes - MDPs, marker dependency pairs. Story 1, Story 2, Story 3

    Paper Tsherniak, Aviad, Francisca Vazquez, Phil G. Montgomery, Barbara A. Weir, Gregory Kryukov, Glenn S. Cowley, Stanley Gill, et al. “[Defining a Cancer Dependency Map](https://doi.org/10.1016/j.cell.2017.06.010).” Cell 170, no. 3 (July 2017)
  • shinyDepMap - a tool to identify targetable cancer genes and their functional connections from Cancer Dependency Map data. Combines CRISPR and shRNA data to determine, for each gene, the growth reduction caused by knockout/knockdown and the selectivity of this effect across cell lines. Two new measures: the degree to which loss of the gene reduces cell growth in sensitive lines (‘efficacy’), and the degree to which its essentiality varies across lines (‘selectivity’). Clusters (tSNE+DBSCAN, robustified with ECHODOTS procedure) genes with similar dependencies across 423 cell lines, revealing functional relationships. Can be used to (1) predict the efficacy and selectivity of drugs targeting particular genes; (2) identify maximally sensitive cell lines for testing a drug; (3) target hop, that is, navigate from an undruggable protein with the desired selectivity profile, such as an activated oncogene, to more druggable targets with a similar profile; and (4) identify novel pathways driving cancer cell growth and survival. Examples of functionality. Data - dependency scores, efficacy, selectivity measures for protein-coding genes across cell lines. GitHub.
    Paper Shimada, Kenichi, John A Bachman, Jeremy L Muhlich, and Timothy J Mitchison. “ShinyDepMap, a Tool to Identify Targetable Cancer Genes and Their Functional Connections from Cancer Dependency Map Data.” ELife 10 (February 8, 2021): e57116. https://doi.org/10.7554/eLife.57116.
  • DGB - Drug Gene Budger, small molecule prioritization using LINCS L1000, CMap, GEO, CREEDS. Output - drugs that up- or downregulated the selected gene, stratified per database.
    Paper Wang, Zichen, Edward He, Kevin Sani, Kathleen M Jagodnik, Moshe C Silverstein, and Avi Ma’ayan. “[Drug Gene Budger (DGB): An Application for Ranking Drugs to Modulate a Specific Gene Based on Transcriptomic Signatures](https://doi.org/10.1093/bioinformatics/bty763).” Edited by Jonathan Wren. Bioinformatics 35, no. 7 (April 1, 2019): 1247–48.
  • DGIdb - drug-gene interaction database integrating 30 sources. API access. Downloads in text format. R/Bioconductor rDGIdb package.
    Paper Cotto, Kelsy C, Alex H Wagner, Yang-Yang Feng, Susanna Kiwala, Adam C Coffman, Gregory Spies, Alex Wollam, Nicholas C Spies, Obi L Griffith, and Malachi Griffith. “[DGIdb 3.0: A Redesign and Expansion of the Drug–Gene Interaction Database](https://doi.org/10.1093/nar/gkx1143).” Nucleic Acids Research, (January 4, 2018)

  • DSigDB - drug-gene signature database. D1 (approved drugs), D2 (kinase inhibitors), D3 (perturbagent signatures), D4 (computational predictions). Download. Yoo, Minjae, Jimin Shin, Jihye Kim, Karen A. Ryall, Kyubum Lee, Sunwon Lee, Minji Jeon, Jaewoo Kang, and Aik Choon Tan. “DSigDB: Drug Signatures Database for Gene Set Analysis” Bioinformatics 31, no. 18 (September 15, 2015)

  • Drug Repurposing Hub - drugs with targets, manually curated, experimentally validated. Data, drugs and targets.

    Paper Corsello, Steven M, Joshua A Bittker, Zihan Liu, Joshua Gould, Patrick McCarren, Jodi E Hirschman, Stephen E Johnston, et al. “[The Drug Repurposing Hub: A next-Generation Drug Library and Information Resource](https://doi.org/10.1038/nm.4306).” Nature Medicine 23, no. 4 (April 2017)

  • GDA - Genomics and Drugs integrated Analysis. The Genomics and Drugs integrated Analysis portal (GDA) is a web-based tool that combines NCI60 uniquely large number of drug sensitivity data with CCLE and NCI60 gene mutation and expression profiles. Gene-to-drug and reverse analysis.

  • OncoKB - OncoKB cancer gene-drug database, different levels of evidence, fully downloadable. API access.

    Chakravarty, Debyani, Jianjiong Gao, Sarah M. Phillips, Ritika Kundra, Hongxin Zhang, Jiaojiao Wang, Julia E. Rudolph, et al. “[OncoKB: A Precision Oncology Knowledge Base](https://doi.org/10.1200/po.17.00011).” JCO Precision Oncology 2017 (July 2017).

  • oncoPharmaDB - R package providing a dataset and method to query targeted and non-targeted cancer drugs, including comprehensive annotations per target, drug mechanism-of-action, approval dates, clinical trial phases for various indications etc.

  • PharmacoDB - A database to mine cancer pharmacogenomics datasets. Guide to the database. Data download and Docker image. GitHub

    Paper Smirnov, Petr, Victor Kofia, Alexander Maru, Mark Freeman, Chantal Ho, Nehme El-Hachem, George-Alexandru Adam, Wail Ba-alawi, Zhaleh Safikhani, and Benjamin Haibe-Kains. “[PharmacoDB: An Integrative Database for Mining in Vitro Anticancer Drug Screening Studies](https://doi.org/10.1093/nar/gkx911).” Nucleic Acids Research, October 9, 2017.

  • PMKB The Precision Medicine Knowledgebase is a project of the Englander Institute for Precision Medicine (EIPM) at Weill Cornell Medicine. API access. Download.

  • TTD (Therapeutic Target Database), contains 1) target-regulating miRNAs and TFs, 2) target-interacting proteins, and 3) patented agents and their targets. Uses ICD-11 codes, support for ICD-9 and ICD-10 remains. Also includes COVID-19 target and drug database. Data downloads, full and subsets. Mirror 2, Mirror 3

    Paper Wang, Yunxia, Song Zhang, Fengcheng Li, Ying Zhou, Ying Zhang, Zhengwen Wang, Runyuan Zhang, et al. “[Therapeutic Target Database 2020: Enriched Resource for Facilitating Research and Early Development of Targeted Therapeutics](https://doi.org/10.1093/nar/gkz981),” Nucleic Acids Research, 06 November 2019
  • Drug combination screen, synergy. Statistics for the analysis of large-scale drug screens, 108 drugs, 40 cell lines. Bliss independence model description. Bliss-based linear model to evaluate viabilities for individual drugs. GitHub. Raw data
    Paper Amzallag, Arnaud, Sridhar Ramaswamy, and Cyril H. Benes. “[Statistical Assessment and Visualization of Synergies for Large-Scale Sparse Drug Combination Datasets](https://doi.org/10.1186/s12859-019-2642-7).” BMC Bioinformatics 20, no. 1 (December 2019).
  • Multi-omics profiling of drug response (90 compounts) in 84 human breast cancer cell lines. RNA-seq is the most predictive, omics measures correlate. DNA copy number (Affymetrix SNP6 - EGA accessions EGAS00000000059 and EGAS00001000585), mRNA expression (Affymetrix U133A and Exon 1.0 ST array - ArrayExpress accessions E-TABM-157 and E-MTAB-181), transcriptome sequence (RNAseq - Gene Expression Omnibus (GEO) accession GSE48216), promoter methylation (Illumina Methylation27 BeadChip - GEO accession GSE42944), protein abundance (Reverse Protein Lysate Array - Additional file 2), and mutation status (Exome-Seq - GEO accession GSE48216). Supplementary material: Supplementary Table 1: Overview of 84 cell lines with subtype information and available data. -log10 (GI50) values for 90 therapeutic compounds are provided for 70/84 cell lines included in all analyses. Supplementary Table 2: Processed Reverse Protein Lysate Array (RPPA) intensity data for 70 (phospho)proteins with fully validated antibodies in 49 cell lines. See Supplementary Methods for data processing details. Supplementary Table 11: Omics signatures for the 22 compounds (ranked genes).
    Paper Daemen, Anneleen, Obi L Griffith, Laura M Heiser, Nicholas J Wang, Oana M Enache, Zachary Sanborn, Francois Pepin, et al. “Modeling Precision Treatment of Breast Cancer.” Genome Biology 14, no. 10 (2013): R110. https://doi.org/10.1186/gb-2013-14-10-r110.

Synergy

  • Review of drug response prediction methods in cancer. Overview of deep learning architectures, Table 1 - DL terminology, Table 2 - data resources (NCI-60, CCLE, GDSC1 etc.), Table 3 - drug combination screening studies. DREAM challenges addressing drug sensitivity. Approaches for building and evaluating drug response prediction models, single-drug (Table 4 - studies, models) and drug combination (Table 6) approaches. DeepSynergy, NCI-ALMANAC dataset, AstraZeneca-Sanger Drug Combination DREAM challenge paper. Evaluation of model ensembles. Deep learning for drug repurposing. Methods for model evaluation, improving performance, increase interpretability. Deep reading with lots of references.
    Paper Baptista, Delora, Pedro G. Ferreira, and Miguel Rocha. "Deep learning for drug response prediction in cancer." Briefings in bioinformatics 22, no. 1 (January 2021): 360-379. https://doi.org/10.1093/bib/bbz171
  • Review of drug synergy methods. Loewe additive model (Dose Equivalence Principle, DEP), Bliss independence model (Multiplicative Survival Principle, MSP), Highest Single Agent (HSA) model. Extensions by Chou and Talalay. Figure 1 - development timeline, the majority of synergy studies do not mention methods. Table 1 - comparison of drug synergy models, references, tools. Minimalistic/optimized sampling schemes. Comparison of SynergyFinder and Combenefit (both use Bliss model), various metrics, DREAM challenge results - results are different, models also diverge for higher-order interactions. GitHub - Code to reproduce analyses.
    Paper Meyer, Christian T., David J. Wooten, Carlos F. Lopez, and Vito Quaranta. “Charting the Fragmented Landscape of Drug Synergy.” Trends in Pharmacological Sciences 41, no. 4 (April 2020): 266–80. https://doi.org/10.1016/j.tips.2020.01.011.
  • Review of drug synergy prediction methods. Description of main models, Loewe additivity, Bliss independence, Combinatorial Index, others. Table 1 - drug synergy studies, techniques. Table 2 - data sources of drug combination screens, public, e.g., NCATS matrix. Table 4 - software, commercial and free.
    Paper Bulusu, Krishna C., Rajarshi Guha, Daniel J. Mason, Richard P.I. Lewis, Eugene Muratov, Yasaman Kalantar Motamedi, Murat Cokol, and Andreas Bender. “Modelling of Compound Combination Effects and Applications to Efficacy and Toxicity: State-of-the-Art, Challenges and Perspectives.” Drug Discovery Today 21, no. 2 (February 2016): 225–38. https://doi.org/10.1016/j.drudis.2015.09.003.
  • BIGL - extension of Loewe model for detecting synergistic effect of drug treatment. Heatmap of combinatorial synergy of main drugs.
    Paper Van der Borght, Koen, Annelies Tourny, Rytis Bagdziunas, Olivier Thas, Maxim Nazarov, Heather Turner, Bie Verbist, and Hugo Ceulemans. “BIGL: Biochemically Intuitive Generalized Loewe Null Model for Prediction of the Expected Combined Effect Compatible with Partial Agonism and Antagonism.” Scientific Reports 7, no. 1 (December 2017). https://doi.org/10.1038/s41598-017-18068-5.
  • DeepSynergy - predicting drug synergy in cancer using neural networks. Uses chemical (three types) and genomic (gene expression, Affy arrays) information as input (concatenated). Uses Merck dataset, 23,062 samples, 583 combinations, 39 human cancer cell lines derived from 7 tiddue types, 38 drugs. a normalization strategy (tanh), and conical layers architecture (two layers, 8192 and 4096 neurons, RELU between layers, linear activation for the output layer, relatively small training rate). Trained on Loewe additive values calculated using Combenefit. Outperforms GBM, RF, SVM, and Elastic Nets as measured by MSE and other metrics (cross-validation). GitHub - Jupyter notebook for data normalization and cross validation. Web interface.
    Paper Preuer, Kristina, Richard PI Lewis, Sepp Hochreiter, Andreas Bender, Krishna C. Bulusu, and Günter Klambauer. "DeepSynergy: predicting anti-cancer drug synergy with Deep Learning." Bioinformatics 34, no. 9 (01 May 2018): 1538-1546. https://doi.org/10.1093/bioinformatics/btx806
  • DCDB - Drug combination database, drugs, combinations, targets. Drug interactions are divided into pharmacodynamic and pharmacokinetic interactions, further subdivided into four types each. 1363 drug combinations, 806 small compounds and 98 biotech drugs, 814 drug targets. Not updated from 2014. Download as plain text.
    Paper Liu, Yanbin, Qiang Wei, Guisheng Yu, Wanxia Gai, Yongquan Li, and Xin Chen. “DCDB 2.0: A Major Update of the Drug Combination Database” 2014 Dec 23, https://doi.org/10.1093%2Fdatabase%2Fbau124
  • DrugCombDb - Combinatorial drug database. Integration of three high-throughput screening studies, literature mining, other databases. Zero Interaction Potency (ZIP) model to quantify synergy/antagonism (SynergyFinder tool). Min-max normalized to 0-1 range.
    Paper Deng, Lei, Bo Zou, Wenhao Zhang, and Hui Liu. “DrugCombDB: A Comprehensive Database of Drug Combinations toward Network Medicine and Combination Therapy,” November 27, 2018. https://doi.org/10.1101/477547.
  • MuSyC - a consensus framework for multi-drug synergy analysis, distinguishes different synergy types (potency, efficacy, cooperativity). A mass-action-based formalism to quantify drug synergy principles and map common frameworks (Dose Equivalence Principle (Loewe), Multiplicative Survival Principle (Bliss), Combinatorial Index (Chou&Talalay), Highest Single Agent, Effective Doze, ZIP, BRAID, Hill PDE, the General Pharmacodynamic Interaction model) onto a unified synergy landscape. Description of each framework (also, Supplementary) and how MuSyC satisfies either under certain conditions. Mixing potency and efficacy masks synergistic interactions. MSP/DEP-based frameworks are biased, the highly cited CI model has two errors. Demonstrated on five drug screening datasets that in some cases synergy predictions are contradictory/wrong (code and data to reproduce are available on BitBucket). Interactive Jupyter notebook on GitHub. Web interface, Supplementary material describes data format and usage.
    Paper Wooten, David J., Christian T. Meyer, Alexander L. R. Lubbock, Vito Quaranta, and Carlos F. Lopez. “MuSyC Is a Consensus Framework That Unifies Multi-Drug Synergy Metrics for Combinatorial Drug Discovery.” Nature Communications 12, no. 1 (December 2021): 4607. https://doi.org/10.1038/s41467-021-24789-z.
  • SynergyFinder - drug synergy analysis using four major reference models (HSA, Loewe, Bliss, ZIP). Web tool and R/Bioconductor package.
    Paper Ianevski, Aleksandr, Liye He, Tero Aittokallio, and Jing Tang. “SynergyFinder: A Web Application for Analyzing Drug Combination Dose-Response Matrix Data.” Bioinformatics (Oxford, England) 33, no. 15 (August 1, 2017): 2413–15. https://doi.org/10.1093/bioinformatics/btx162.
  • Synergy interactive heatmap of a large drug combination screen (108 drugs, 40 cell lines, two concentrations). Bliss independence model description, statistics. Bliss-based linear model (log of the Bliss model) to evaluate viabilities for individual drugs. GitHub - Code to reproduce the analysis. CSV - Raw data.
    Paper Amzallag, Arnaud, Sridhar Ramaswamy, and Cyril H. Benes. “Statistical Assessment and Visualization of Synergies for Large-Scale Sparse Drug Combination Datasets.” BMC Bioinformatics 20, no. 1 (December 2019). https://doi.org/10.1186/s12859-019-2642-7.

Tools

Preprocessing

  • The Trinity Cancer Transcriptome Analysis Toolkit (CTAT) aims to provide tools for leveraging RNA-Seq to gain insights into the biology of cancer transcriptomes. Bioinformatics tool support is provided for mutation detection, fusion transcript identification, de novo transcript assembly of cancer-specific transcripts, lncRNA classification, and foreign transcript detection (viruses, microbes). ctat-mutations - Mutation detection using GATK4 best practices and latest RNA editing filters resources. Works with both Hg38 and Hg19

  • cacao - Callable Cancer Loci - assessment of sequencing coverage for actionable and pathogenic loci in cancer, example of QC report. Data: BED files for cancer loci from ClinVar, CIViC, cancerhotspots.

Purity

  • GenomeScope and Smudgeplot for ploidy detection directly from sequencing data. Based on k-mer counting using KMC or Jellyfish, negative binomial-based mathematical model. Web-GenomeScope.
    Paper Ranallo-Benavidez, T Rhyker, Kamil S Jaron, and Michael C Schatz. “GenomeScope 2.0 and Smudgeplot for Reference-Free Profiling of Polyploid Genomes,” Nature Communications volume 11, Article number: 1432 (2020) https://doi.org/10.1038/s41467-020-14998-3

  • ABSOLUTE - infers tumor purity, ploidy from SNPs, CNVs. Also detects subclonal heterogeneity.

    • Carter, Scott L., Kristian Cibulskis, Elena Helman, Aaron McKenna, Hui Shen, Travis Zack, Peter W. Laird, et al. “Absolute Quantification of Somatic DNA Alterations in Human Cancer.” Nature Biotechnology 30, no. 5 (May 2012): 413–21. https://doi.org/10.1038/nbt.2203.
    • Aran, Dvir, Marina Sirota, and Atul J. Butte. “Systematic Pan-Cancer Analysis of Tumour Purity.” Nature Communications 6, no. 1 (December 2015). https://doi.org/10.1038/ncomms9971. - TCGA tumor purity estimation using four methods: ESTIMATE, ABSOLUTE, LUMP, IHC, and a median consensus purity estimation. Gene expression correlates with purity and may affect correlation and differential expression detection analyses - big confounding effect.
      • data/ABSOLUTE_scores.xlsx - Supplementary Data 1: Tumor purity estimates according to four methods and the consensus method for all TCGA samples with available data. Source

  • ESTIMATE (Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data) is a tool for predicting tumor purity, and the presence of infiltrating stromal/immune cells in tumor tissues using gene expression data. ESTIMATE algorithm is based on single sample Gene Set Enrichment Analysis and generates three scores: stromal score (that captures the presence of stroma in tumor tissue), immune score (that represents the infiltration of immune cells in tumor tissue), and estimate score (that infers tumor purity). http://bioinformatics.mdanderson.org/main/ESTIMATE:Overview. R package http://bioinformatics.mdanderson.org/estimate/rpackage.html

    • Yoshihara, Kosuke, Maria Shahmoradgoli, Emmanuel Martínez, Rahulsimham Vegesna, Hoon Kim, Wandaliz Torres-Garcia, Victor Treviño, et al. “Inferring Tumour Purity and Stromal and Immune Cell Admixture from Expression Data.” Nature Communications 4 (2013): 2612. https://doi.org/10.1038/ncomms3612. - ESTIMATE - tumor-stroma purity detection. 141 immune and stromal genes. single-sample GSEA analysis. ESTIMATE score as a combination of immune and stromal scores. Supplementary data.

  • data/ESTIMATE_signatures.xlsx - A gene list of stromal and immune signatures. Source

  • data/ESTIMATE_scores.xlsx - A list of stromal, immune, and ESTIMATE scores in TCGA data sets. All cancers, all gene expression plaforms. Source

  • ISOpureR - Deconvolution of Tumour Profiles to purify tumor samples. Regression-based, uses purified tumor profile to estimate the proportion of tumor samples. Discussion of overfitting due to overparametrization

Ploidy

  • GenomeScope and Smudgeplot for ploidy detection directly from sequencing data. Based on k-mer counting using KMC or Jellyfish, negative binomial-based mathematical model. GenomeScope.
    Paper Ranallo-Benavidez, T. Rhyker, Kamil S. Jaron, and Michael C. Schatz. "GenomeScope 2.0 and Smudgeplot for reference-free profiling of polyploid genomes." Nature Communications 11, no. 1 (2020): 1-10. https://doi.org/10.1038/s41467-020-14998-3
  • Sequenza - paired tumor-normal WGS and WES analysis. Estimates tumor cellularity, ploidy, infer allele-speific copy number profiles. Python-based sequenza-utils perform data preprocessing. R package sequenza performs the analysis. Documentation.
    Paper Favero, F., T. Joshi, A. M. Marquard, N. J. Birkbak, M. Krzystanek, Q. Li, Z. Szallasi, and A. C. Eklund. “Sequenza: Allele-Specific Copy Number and Mutation Profiles from Tumor Sequencing Data.” Annals of Oncology 26, no. 1 (January 2015): 64–70. https://doi.org/10.1093/annonc/mdu479.

Immune cell deconvolution

  • DeMixT - tumor-stroma-immune transcriptome deconvolution using a three-component model. Requires reference profiles from the stromal component. Methods, the iterated conditional modes (ICM) algorithm and a gene-set-based component merging (GSCM) approach for parameter estimation, adaptive integration. Estimates proportions and reconstructs patient-speciric gene expression matrices. More accurate than ISOpure, similar to CIBERSORT. R package, parallelization with OpenMP.

    Paper Wang, Zeya, Shaolong Cao, Jeffrey S. Morris, Jaeil Ahn, Rongjie Liu, Svitlana Tyekucheva, Fan Gao, et al. “Transcriptome Deconvolution of Heterogeneous Tumor Samples with Immune Infiltration.” IScience 9 (November 2018): 451–60. https://doi.org/10.1016/j.isci.2018.10.028.

    Cao, Shaolong, Zeya Wang, Fan Gao, Jingxiao Chen, Feng Zhang, Daniel E Frigo, Eleni Efstathiou, Scott Kopetz, and Wenyi Wang. “An R Implementation of Tumor-Stroma-Immune Transcriptome Deconvolution Pipeline Using DeMixT.” BioRxiv, January 1, 2019, 566075. https://doi.org/10.1101/566075.

  • Tumor Immune Single-cell Hub (TISCH) is a scRNA-seq database focusing on tumor microenvironment (TME). TISCH provides detailed cell-type annotation at the single-cell level, enabling the exploration of TME across different cancer types. Tweet

  • quanTIseq - quantification of the Tumor Immune cell proportions from human RNA-seq data. Input - FASTQ files (Trimmomatic, Kallisto to TPM, normalization), or TPM matrix. Deconvolution into 10 cell types (B, macrophages M1 and M2, Monocytes, Neutrophils, NK, CD8 T, CD4 T, Dendritic cells), and uncharacterized fraction (TIL10 signature). Custom processing of 51 datasets to generate TIL10. Compared with CIBERSORT, TIMER, EPIC on simulated and real-life data. Command line, Docker/Singularity implementation, no GitHub.

  • CIBERSORT - cell type identification using Support Vector Regression. p-value for the overall goodness of deconvolution (H0 - no cell types are present in a given gene expression profile), also Pearson and RMSE for estimating goodness of fit. References to six GEP deconvolution methods: linear least-squares regression (LLSR), quadratic programming (QP), perturbation model for gene expression deconvolution (PERT), robust linear regression (RLR), microarray microdissection with analysis of differences (MMAD) and digital sorting algorithm (DSA). References to datasets for benchmarking.

  • DeconRNAseq - deconvolution of RNA-seq datasets into cell proportions using cell signatures. Non-negative decomposition algorithm (X = AS) solved using quadratic programming

  • Immunophenogram - partitioning immune cell types in cancer. https://github.com/mui-icbi/Immunophenogram, https://tcia.at/home

    • Charoentong, Pornpimol, Francesca Finotello, Mihaela Angelova, Clemens Mayer, Mirjana Efremova, Dietmar Rieder, Hubert Hackl, and Zlatko Trajanoski. “Pan-Cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade.” BioRxiv, 2016, 056101. - https://tcia.at/ - immune cells in cancers. Estimated using functional GSEA enrichment, and CIBERSORT. Immunophenogram generation: https://github.com/MayerC-imed/Immunophenogram

  • ImmQuant - Deconvolution of immune cell lineages. http://csgi.tau.ac.il/ImmQuant/downloads.html

    • Frishberg, Amit, Avital Brodt, Yael Steuerman, and Irit Gat-Viks. “ImmQuant: A User-Friendly Tool for Inferring Immune Cell-Type Composition from Gene-Expression Data.” Bioinformatics 32, no. 24 (December 15, 2016): 3842–43. https://doi.org/10.1093/bioinformatics/btw535.

  • TIMER - a resource to estimate the abundance of immune infiltration of six cell types, the effect on survival. Accounting for tumor purity using CHAT R package. Linear regression to estimate immune cell abundance. Macrophage infiltration predicts worse outcome, including BRCA. Signature genes from microarray, merged with RNA-seq using ComBat for batch removal, filtered. All TCGA processed. All data at http://cistrome.org/TIMER/download.html, tool at https://cistrome.shinyapps.io/timer/

    • Li, Bo, Eric Severson, Jean-Christophe Pignon, Haoquan Zhao, Taiwen Li, Jesse Novak, Peng Jiang, et al. “Comprehensive Analyses of Tumor Immunity: Implications for Cancer Immunotherapy.” Genome Biology 17, no. 1 (22 2016): 174. https://doi.org/10.1186/s13059-016-1028-7.

  • TIDE - Tumor Immune Dysfunction and Exclusion, a gene expression biomarker to predict the clinical response to immune checkpoint blockade using patient-specific gene expression. http://tide.dfci.harvard.edu/

  • TRUST4 - infers tumor-infiltrating TCR and BCR repertores from bulk and scRNA-seq (5' 10X Genomics kit). https://github.com/liulab-dfci/TRUST4

BRCA

OvCa

  • PrOTYPE - Ovarian Cancer subtype prediction. Model trained on gene expression from 1650 tumors (specimens from the Ovarian Tumor Tissue Analysis consortium), validated on NanoString data on 3829 tumors. 55 genes, predict with >95% accuracy, also associated with age, stage, residual disease, TILs, outcome. Review of previous studies classifying into 5 outcomes. Random Forest, cross-validation. Supplemental Table SC7 lists all predictor genes. PrOTYPE web tool to classify NanoString OvCa samples, https://ovcare.shinyapps.io/PrOType/
    • Talhouk, Aline, Joshy George, Chen Wang, Timothy Budden, Tuan Zea Tan, Derek S Chiu, Stefan Kommoss, et al. “Development and Validation of the Gene-Expression Predictor of High-Grade-Serous Ovarian Carcinoma Molecular SubTYPE (PrOTYPE).” Clinical Cancer Research, June 17, 2020, clincanres.0103.2020. https://doi.org/10.1158/1078-0432.CCR-20-0103.

SCLC

TCGA

Integrative

  • Review of tools and methods for the integrative analysis of multiple omics data, cancer-oriented. Table 1 - multi-omics data repositories (TCGA, CPTAC, ICGC, CCLE, METABRIC, TARGET, Omics Discovery Index). Three broad areas of multi-omics analysis: 1. Disease subtyping and classification based on multi-omics profiles; 2. Prediction of biomarkers for various applications including diagnostics and driver genes for diseases; 3. Deriving insights into disease biology. Table 2 - software categorized by use case (PARADIGM, iClusterPlus, PSDF, BCC, MDI, SNF, PFA, PINSPlus, NEMO, mixOmics, moCluster, MCIA, JIVE, MFA, sMBPLS, T-SVD, Joint NMF). Brief description of each tool, links, exemplary publications. Table 3 - visualization portals (cBioPortal, Firebrowse, UCSC Xena, LinkedOmics, 3Omics, NetGestalt, OASIS, Paintomics, MethHC). Description of each, data types, analysis examples.

  • CTGS - web portal for Cancer Target Gene Screening. Visualization, survival. Gene expression, methylation, CNAs, SNPs, clinical info. METABRIC, TCGA, other published data. Similar services: cBioPortal, Breast Cancer Integrated Platform (BCIP), MOBCdb, Oncomine, Kaplan–Meier (KM) plotter and TCGA4U.

    Paper Kim, Hyung-Yong, Hee-Joo Choi, Jeong-Yeon Lee, and Gu Kong. “Cancer Target Gene Screening: A Web Application for Breast Cancer Target Gene Screening Using Multi-Omics Data Analysis.” Briefings in Bioinformatics 21, no. 2 (March 23, 2020): 663–75. https://doi.org/10.1093/bib/bbz003.
  • LinkedOmics - clinical, gennomic (expression, SNPs, CNVs, methylation, miRNAs) and protein expression data from TCGA, CPTAC. Three analysis modules: LinkFinder finds associations between molecular and clinical attributes; LinkCompare compares the associations. LinkInterpreter maps associations to pathways and networks.

Deep Learning

  • TDC (Therapeutic Data Commons) - Kaggle for therapeutics, discovery and development of safe and effective medicines. A framework to evaluate machine learning, contains 66 datasets, 22 learning tasks (single-instance, multi-instance, generative learning). PyTDC - python interface. The paper describes details of each data, task, resource.

  • Review of drug response prediction methods in cancer. Overview of deep learning architectures, Table 1 - DL terminology, Table 2 - data resources (NCI-60, CCLE, GDSC1 etc.), Table 3 - drug combination screening studies. DREAM challenges addressing drug sensitivity. Approaches for building and evaluating drug response prediction models, single-drug (Table 4 - studies, models) and drug combination (Table 6) approaches. DeepSynergy, NCI-ALMANAC dataset, AstraZeneca-Sanger Drug Combination DREAM challenge paper. Evaluation of model ensembles. Deep learning for drug repurposing. Methods for model evaluation, improving performance, increase interpretability. Deep reading with lots of references.

  • DrugCell - interpretable neural network, directly mapping neurons of a deep neural network to Gene Ontology hierarchy. Input - genotype (binary gene indicator) and drugs, output - response of cell to drug. Performance is the same as unconstrained network. Trained on 1235 tumor cell lines and 684 drugs. Allows for the detection of synergistic combinations. Perspective

  • Combined neoantigen burden and immune-related resistance mechanisms predict immunotherapy response（https://codeocean.com/capsule/1428816/tree/v1）

  • Immunogenomic analysis of pediatric nervous system tumors（https://codeocean.com/capsule/9176641/tree/v1）

  • A global and integrated analysis of CINSARC-associated genetic defects（https://codeocean.com/capsule/4933686/tree/v4）

  • Drug response prediction from gene expression data. Deep Neural Network (DNN, H2O.ai framework) compared with Elastic Net, Random Forest. Trained on highly variable (by MAD) gene expression in 1001 cell lines and 251 drugs pharmacogenomic dataset (GDSC. CCLP) to predict IC50. Hyper-parameter optimization using 5-fold cross-validation and minimizing Mean Square Error. Batch correction between the datasets Tested on unseen patient cohorts (OCCAMS, MD Anderson, TCGA, Multiple Myeloma Consortium) to predict IC50 and test low, medium, high IC50 groups for survival differences. RDS files data, R code

Image analysis

  • HISTOBREAST - brightfield microscopy images of Haematoxylin - Eosin (H&E) stained breast tissue (moderately differentiated (G2) invasive breast cancer of no special type). Different acquisition conditions, two magnifications (5X, 50X). Neighbour image tiles (16) and ensemble of mosaics composed from different combinations of the available image tiles, exhibiting progressively degraded quality levels. Can serve for benchmarking and developing new image processing techniques.
    Paper Buga, Roxana M., Tiberiu Totu, Adrian Dumitru, Mariana Costache, Iustin Floroiu, Nataša Sladoje, and Stefan G. Stanciu. “HISTOBREAST, a Collection of Brightfield Microscopy Images of Haematoxylin and Eosin Stained Breast Tissue.” Scientific Data 7, no. 1 (December 2020): 169. https://doi.org/10.1038/s41597-020-0500-0.

Clonal analysis

  • Awesome-CancerEvolution - list of papers and tools for studying cancer evolution. https://github.com/iron-lion/Awesome-CancerEvolution

  • Single-cell lineage-tracing technology, to study the routes and drives of metastasis. Cas9 cuts a defined genomic locus resulting in a stable indel allele; as cell divide, they accrue more indels at additional sites that will be used to resolve phylogenies of individual cells. Engineered A549 cells (A549-LT) to contain luciferase for live imaging, Cas9 for generating heritable indels, 10 uniquely barcoded copies of the target site to record lineage information, sgRNAs to direct Cas9 to target sites. Embedded in matrigel, implanted in ling of immunodeficient mice. Out of 5000 cells, 2100 clones can be distinguished, only about 100 engrafted. Investigation of metastatic drivers, heterogeneity. Phylogenies reconstructed using Cassiopeia. Supplementary material - Tables of metastasis-associated genes identified in the four mouse experiments, including positive and negative metastatic gene signature.

    Paper Quinn, Jeffrey J., Matthew G. Jones, Ross A. Okimoto, Shigeki Nanjo, Michelle M. Chan, Nir Yosef, Trever G. Bivona, and Jonathan S. Weissman. “Single-Cell Lineages Reveal the Rates, Routes, and Drivers of Metastasis in Cancer Xenografts.” Science 371, no. 6532 (February 26, 2021): eabc1944. https://doi.org/10.1126/science.abc1944.
  • HATCHet (Holistic Allele-specific Tumor Copy- number Heterogeneity) -  a Python tool that infers allele- and clone-specific CNAs and WGDs jointly across multiple tumor samples from the same patient. Outperforms six methods (Battenberg, TITAN, THetA, cloneHD, Canopy, and ReMixT) on multi-sample DNA sequencing data. MASCoTE (Multiple Allele-specific Simulation of Copy-number Tumor Evolution) tool for data simulation. Input: Read-depth ratio (RDR) and B-allele frequency (BAF) for short genomic bins across samples from the same patient. Output: copy number states and clone proportions. Code to reproduce the paper.
    Paper Zaccaria, Simone, and Benjamin J. Raphael. “Accurate Quantification of Copy-Number Aberrations and Whole-Genome Duplications in Multi-Sample Tumor Sequencing Data.” Nature Communications 11, no. 1 (December 2020): 4301. https://doi.org/10.1038/s41467-020-17967-y.

Survival analysis

  • cSurvival - multi-omics survival analysis. Expression, mutation, CNV, miRNA, methylation, protein measures. Optimal cutoff selection by scanning for minimal p-value, grid search for multiple predictors, correction for multiple testing. Differential dependency and cell viability analysis. TCGA, TARGET, DepMap data. eVITTA gene signatures. Three additional features: (i) joint analysis with two genomic predictors to identify interacting biomarkers; (ii) gene set level analysis; and (iii) integration of clinical and experimental cell line studies to generate synergistic biological insights.
    Paper Cheng, Xuanjin, Yongxing Liu, Jiahe Wang, Yujie Chen, Andrew Gordon Robertson, Xuekui Zhang, Steven J M Jones, and Stefan Taubert. “CSurvival: A Web Resource for Biomarker Interactions in Cancer Outcomes and in Cell Lines.” Briefings in Bioinformatics, April 2, 2022, bbac090. https://doi.org/10.1093/bib/bbac090.

  • www.tcgaportal.org - web server for survival analysis using TCGA data

  • cBioPortal - The cBioPortal for Cancer Genomics provides visualization, analysis and download of large-scale cancer genomics data sets. OncoPrint mutation plots, differential expression, coexpression, survival. Compare gene expression with copy number variation. http://www.cbioportal.org/

  • R2 - Genomics Analysis and Visualization Platform. Gene-centric, survival analysis, collection of preprocessed microarray studies. http://hgserver1.amc.nl/

  • KM plotter - Gene-centric, customizable survival analysis for breast, ovarian, lung, gastric cancers. http://kmplot.com/

    • Györffy, Balazs, Andras Lanczky, Aron C. Eklund, Carsten Denkert, Jan Budczies, Qiyuan Li, and Zoltan Szallasi. “An Online Survival Analysis Tool to Rapidly Assess the Effect of 22,277 Genes on Breast Cancer Prognosis Using Microarray Data of 1,809 Patients.” Breast Cancer Research and Treatment 123, no. 3 (October 2010): 725–31. https://doi.org/10.1007/s10549-009-0674-9.

  • The Human Protein Atlas: Pathology atlas - Gene- and protein expression data in multiple cancer tissues, cell lines. Easy one-gene search, summary of tissue-specific expression, survival significance. http://www.proteinatlas.org/

    • Uhlen, Mathias, Cheng Zhang, Sunjae Lee, Evelina Sjöstedt, Linn Fagerberg, Gholamreza Bidkhori, Rui Benfeitas, et al. “A Pathology Atlas of the Human Cancer Transcriptome.” Science (August 18, 2017). Data download - tissue-specific gene expression in cancer and normal, isoform expression, protein expression. Supplementary material
      • Table S2 - summary of tissue specific expression for each gene, in normal and cancer tissues.
      • Table S6 - summary of survival prognostic value, with a simple "favorable/unfavorable" label for each gene. Each worksheet corresponds to a different cancer.
      • Table S8 - per-gene summary, in which cancers it is prognostic of survival.

  • Breast Cancer Gene-Expression Miner v4.4 - gene expression, correlation, and survival analysis in different microarray (e.g., METABRIC) and RNA-seq (e.g., TCGA) datasets

  • G-2-O, Genotype to Outcome - web-server linking mutation (or CNV) of a gene to clinical outcome (survival) by utilizing next generation sequencing and gene chip data. For Breast and Lung cancer

  • PRECOG - PREdiction of Clinical Outcomes from Genomic Profiles. Gene-centric, quick overview of survival effect of a gene across all cancers, KM plots. https://precog.stanford.edu

    • Gentles, Andrew J., Aaron M. Newman, Chih Long Liu, Scott V. Bratman, Weiguo Feng, Dongkyoon Kim, Viswam S. Nair, et al. “The Prognostic Landscape of Genes and Infiltrating Immune Cells across Human Cancers.” Nature Medicine 21, no. 8 (August 2015): 938–45. https://doi.org/10.1038/nm.3909. - TCGA pan-cancer survival analysis PRECOG, CIBERSORT. 39 cancers. Intro into heterogeneity. Z-score description. Batch effect does not significantly affect z-scores. 2/3 prognostic genes shared across cancers. AutoSOME clustering method

  • GEPIA - single- and multiple-gene analyses of TCGA data. Gene expression in different tumor-normal comparisons, differentially expressed genes, correlation analysis, similar genes, survival analysis. http://gepia.cancer-pku.cn/

    • Zefang Tang et al., “GEPIA: A Web Server for Cancer and Normal Gene Expression Profiling and Interactive Analyses,” Nucleic Acids Research 45, no. W1 (July 3, 2017): W98–102, https://doi.org/10.1093/nar/gkx247. - TCGA and GTEX web interface. Classical analyses - differential expression analysis, profiling plotting, correlation analysis, patient survival analysis, similar gene detection and dimensionality reduction analysis. http://gepia.cancer-pku.cn/

  • GEPIA2 - isoform-level TCGA analysis. Cancer subtype-specific analyses. Eight types of expression analyses, and additional Cancer Subtype Classifier and Expression Comparison. Python package for API access. http://gepia2.cancer-pku.cn

    • Tang, Zefang, Boxi Kang, Chenwei Li, Tianxiang Chen, and Zemin Zhang. “GEPIA2: An Enhanced Web Server for Large-Scale Expression Profiling and Interactive Analysis.” Nucleic Acids Research, May 22, 2019. https://doi.org/10.1093/nar/gkz430.

  • UALCAN - Gene-centric, tumor-normal expression, survival analusis, TCGA cancers. http://ualcan.path.uab.edu/

    • Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Rodriguez IP, Chakravarthi BVSK and Varambally S. UALCAN: A portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia. 2017 Aug;19(8):649-658. doi: 10.1016/j.neo.2017.05.002 [PMID:28732212]

  • Project Betastasis - Gene-centric, survival analysis, gene expression, select cancer studies. http://www.betastasis.com/

  • OncoLnc - Gene-centric, survival analysis in any TCGA cancer. http://www.oncolnc.org/

Methods to find best cutoff for survival

  • KMplotter - the Kaplan Meier plotter is capable to assess the effect of 54,675 genes on survival using 18,674 cancer samples. These include 5,143 breast, 1,816 ovarian, 2,437 lung, 364 liver, 1,065 gastric cancer patients with relapse-free and overall survival data. The miRNA subsystems include additional 11,456 samples from 20 different cancer types. Primary purpose of the tool is a meta-analysis based biomarker assessment.

    • Györffy, Balazs, Andras Lanczky, Aron C. Eklund, Carsten Denkert, Jan Budczies, Qiyuan Li, and Zoltan Szallasi. “An Online Survival Analysis Tool to Rapidly Assess the Effect of 22,277 Genes on Breast Cancer Prognosis Using Microarray Data of 1,809 Patients.” Breast Cancer Research and Treatment 123, no. 3 (October 2010): 725–31. https://doi.org/10.1007/s10549-009-0674-9. - cutoff selection for survival by scanning gene expression range.

  • ctree function for automatic cutoff finding and building a regression tree out of multiple covariates. partykit::ctree().

    • Hothorn, Torsten, Kurt Hornik, and Achim Zeileis. “Ctree: Conditional Inference Trees.” The Comprehensive R Archive Network, 2015, 1–34.

  • Cutoff Finder - web tool for finding optimal dichotomization with respect to an outcome or survival variable. Five methods. http://molpath.charite.de/cutoff/

    • Budczies, Jan, Frederick Klauschen, Bruno V. Sinn, Balázs Győrffy, Wolfgang D. Schmitt, Silvia Darb-Esfahani, and Carsten Denkert. “Cutoff Finder: A Comprehensive and Straightforward Web Application Enabling Rapid Biomarker Cutoff Optimization.” PloS One 7, no. 12 (2012): e51862. https://doi.org/10.1371/journal.pone.0051862.

Cancer driver genes

  • geneOncoX - Human gene annotations for the oncology domain, by Sigve Nakken. Integration of multiple cancer gene annotation resources. Tutorial

  • Many cancer genes switch between one-hit and two-hit drivers. Mutations of genes in the same biological pathway is a contributing factor. Higher-order interactions are abundant. GitHub, Supplsmentary Information. Park_2021_cancer_genes.xlsx - Supplementary Dataset 1, 201 genes include 117 tumor-suppressor genes (TSGs), 77 oncogenes (OGs) and 7 dual-function genes (DFGs)

    Paper Park, Solip, Fran Supek, and Ben Lehner. “[Higher Order Genetic Interactions Switch Cancer Genes from Two-Hit to One-Hit Drivers](https://doi.org/10.1038/s41467-021-27242-3).” Nature Communications, (December 2021)
  • CR2Cancer - chromatin regulator (CR) and cancer association database. Genomic, transcriptomic, proteomic, clinical, and functional information for over 400 CRs. Seven categories of knowledge, six view panels ‘Cancer Type’, ‘Function’, ‘Mutation Rate’, ‘Differential Expres- sion’, ‘ChIP-Seq Data’ and ‘Targeted Drug’. Download various CR data.
    Paper Ru, Beibei, Jianlong Sun, Yin Tong, Ching Ngar Wong, Aditi Chandra, Acacia Tsz So Tang, Larry Ka Yue Chow, Wai Lam Wun, Zarina Levitskaya, and Jiangwen Zhang. “CR2Cancer: A Database for Chromatin Regulators in Human Cancer.” Nucleic Acids Research 46, no. D1 (January 4, 2018): D918–24. https://doi.org/10.1093/nar/gkx877.

  • Nucleotide context of mutations is associated with driver/passenger status. Pan-cancer analysis (data from 87 studies, including TCGA), seven methods for driver-gene detection, 460 driver genes clustered into 21 cancer-related pathways. Apoptosis regulation and chromatin modification are recurrent pathways. MutPanning software (Mac, Windows, Java) for analyzing nucleotide content.

    • Dietlein, Felix, Donate Weghorn, Amaro Taylor-Weiner, André Richters, Brendan Reardon, David Liu, Eric S. Lander, Eliezer M. Van Allen, and Shamil R. Sunyaev. “Identification of Cancer Driver Genes Based on Nucleotide Context.” Nature Genetics, February 3, 2020.
    • Dietlein_2020_Drivers.xlsx - Supplementary Tables 1–5
      • Supplementary Table 3 - Stratification of gene-tumor pairs based on their literature support (460 genes aggregated by cancer type).
      • Supplementary Table 4 - Stratification of driver genes based on their literature support (460 genes aggregated by gene).
      • Supplementary Table 5 - Literature references for additional cancer genes (30 genes).

  • MOMA Oncogenic Architecture - A network-based integrative genomic analysis of 20 The Cancer Genome Atlas cohorts characterizes conserved master regulator blocks underlying cancer hallmarks across different tumor types, providing insights into the connection between genetic alterations and tumor transcriptional identity. Tumor Subtypes Explorer

    • Integrative genomic analysis of 20 TCGA cohorts identifies 112 distinct tumor subtypes
    • 407 master regulators (MRs) canalize the effects of mutations to implement cancer states
    • 24 conserved master regulator blocks regulate cancer hallmarks across tumors
    • Paull, Evan O., Alvaro Aytes, Sunny J. Jones, Prem S. Subramaniam, Federico M. Giorgi, Eugene F. Douglass, Somnath Tagore, et al. “A Modular Master Regulator Landscape Controls Cancer Transcriptional Identity.” Cell, (January 2021) - supplementary material with data
      • Table S1 - summary, cancer subtypes.
      • Table S2 - master regulators, tumor checkpoints (hyperconnected modules)
      • Table S4 - 24 MR modules (genes in them), their association with survival, enrichment in hallmarks of cancer, upstream genomics
      • Table S6 - Cluster maps of each cancer cohort into subtypes by master regulators

  • The Network of Cancer Genes - NCG contains information on duplicability, evolution, protein-protein and microRNA-gene interaction, function, expression and essentiality of 2,372 cancer genes from 273 manually curated publications. Downloads

  • The list of 727 cancer genes from whole-genome sequences of 560 breast cancers from Nik-Zainal 2016 paper, Paper

  • The list of cancer-related genes from the Bushman Lab. allOnco_May2018.tsv

  • Integrative pathway and network analysis of 2583 cancers (27 tumor types) identified 87 driver Pathway Implicated Driver (PID) genes with coding variants (PID-C) and 93 drivers with noncoding variants (PID-N). These gene classes are associated with different biological processes. Six pathway databases, seven pathway and network methods, data references in Methods. Non-Coding Added Value (NCVA) score to identify genes with noncoding variants increasing the overall significance.

  • MoonlightR - integrative analysis of TCGA data to predict cancer driver genes. http://bioconductor.org/packages/release/bioc/vignettes/MoonlightR/inst/doc/Moonlight.html, https://github.com/ibsquare/MoonlightR

    • Supplementary Data 5 - Cancer Driver Genes for TCGA BRCA molecular subtypes
    • Supplementary Data 6 - Moonlight’s oncogenic mediators in 18 cancer types
    • Other supplementary data - oncogenic mediators overlapping with methylation, chromatin accessibility, copy number changes, mutations, survival.
    • Colaprico, Antonio, Catharina Olsen, Matthew H. Bailey, Gabriel J. Odom, Thilde Terkelsen, Tiago C. Silva, André V. Olsen, et al. “Interpreting Pathways to Discover Cancer Driver Genes with Moonlight.” Nature Communications 11, no. 1 (December 2020): 69. https://doi.org/10.1038/s41467-019-13803-0.

  • CancerGeneNet - CancerGeneNet is a resource that aims at linking genes that are frequently mutated in cancers to cancer phenotypes. The resource takes advantage of a curation effort aimed at embedding a large fraction of the gene products that are found altered in cancers in the cell network of causal protein relationships. Graph algorithms, in turn, allow to infer likely paths of causal interactions linking cancer associated genes to cancer phenotypes thus offering a rational framework for the design of strategies to revert disease phenotypes. CancerGenNet bridges two interaction layers by connecting proteins whose activities are affected by cancer gene products to proteins that impact on cancer phenotypes. This is achieved by implementing graph algorithms that allow searching for graph path that link any gene of interest to the “hallmarks of cancer". https://signor.uniroma2.it/CancerGeneNet/

  • Breast cancer genes with actionable mutations according to ESMO Scale for Clinical Actionability of molecular Targets (ESCAT). Tiers I-V, X, plus A/B/C letters for level of evidence (Table 1). Tables 2-4 list actual genes and their description.

    Paper Condorelli, R. “Genomic Alterations in Breast Cancer: Level of Evidence for Actionability According to ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT).” Annals of Oncology 30, no. 3 (2019): 9.

  • Cancer Gene Census (CGC), download COSMIC

    • Hudson, T. J. et al. International network of cancer genome projects. Nature 464, 993–8 (2010).
    • data/Census_all.csv - The cancer Gene Census. Updated 2020-12-03
    • data/COSMIC_genes.txt - Genes sorted by the number of records associated with them. Obtained using cat Census_allThu_Dec_3_19_52_03_2020.csv | sed '1d' | cut -f1 -d, | sort | uniq > COSMIC_genes.txt. Updated 2020-12-03
    • data/CosmicCodingMuts.vcf.gz - VCF file of all coding mutations in the current release (release v83, 7th November 2017).

  • Oncology Model Fidelity Score based on the Hallmarks of Cancer, an R/Shiny app to check cancer samples from preprocessed or user-supplied gene expression data for the presence of these hallmarks. https://github.com/tedgoldstein/hallmarks

  • Tumor suppressor gene database (TSGene), https://bioinfo.uth.edu/TSGene/

    • Zhao, M., Sun, J. & Zhao, Z. TSGene: a web resource for tumor suppressor genes. Nucleic Acids Res, 41(Database issue), D970–6 (2013).
    • Download various lists of tumor suppressor genes, https://bioinfo.uth.edu/TSGene/download.cgi

  • OncoScore - an R package, text-mining tool that ranks genes according to their association with cancer, based on available biomedical literature. Outperforms GeneRanker.

    Paper Piazza, Rocco, Daniele Ramazzotti, Roberta Spinelli, Alessandra Pirola, Luca De Sano, Pierangelo Ferrari, Vera Magistroni, Nicoletta Cordani, Nitesh Sharma, and Carlo Gambacorti-Passerini. “OncoScore: A Novel, Internet-Based Tool to Assess the Oncogenic Potential of Genes.” Scientific Reports 7, no. 1 (May 3, 2017): 46290. https://doi.org/10.1038/srep46290.

  • OncoScape - Genes with oncogenic/tumor suppressor/combined scores as a sum contribution from gene expression, somatic mutations, DNA copy-number and methylation as well as data from shRNA knock-down screens. http://oncoscape.nki.nl/

    • Schlicker, Andreas, Magali Michaut, Rubayte Rahman, and Lodewyk F. A. Wessels. “OncoScape: Exploring the Cancer Aberration Landscape by Genomic Data Fusion.” Scientific Reports 6 (20 2016): 28103. https://doi.org/10.1038/srep28103.

  • data/Bailey_2018_cancer_genes.xlsx - Table S1, consensus list of cancer driver genes.

    • Bailey, Matthew H., Collin Tokheim, Eduard Porta-Pardo, Sohini Sengupta, Denis Bertrand, Amila Weerasinghe, Antonio Colaprico, et al. “Comprehensive Characterization of Cancer Driver Genes and Mutations.” Cell 173, no. 2 (April 5, 2018): 371-385.e18. https://doi.org/10.1016/j.cell.2018.02.060. - Pan-Cancer mutation analysis. Combined use of 26 tools (https://www.cell.com/cell/fulltext/S0092-8674(18)30237-X#secsectitle0075, description of each tool in Methods) on harmonized data. 299 cancer driver genes, >3,400 putative missense driver mutations. Table S6 - excluded TCGA samples.

  • data/TARGET_db_v3_02142015.xlsx - TARGET (tumor alterations relevant for genomics-driven therapy) is a database of genes that, when somatically altered in cancer, are directly linked to a clinical action. TARGET genes may be predictive of response or resistance to a therapy, prognostic, and/or diagnostic. https://software.broadinstitute.org/cancer/cga/target

  • data/Tokheim_2016_cancer_driver_genes.xlsx - Dataset S2: Predicted driver genes by various number of methods

    • Tokheim, Collin J., Nickolas Papadopoulos, Kenneth W. Kinzler, Bert Vogelstein, and Rachel Karchin. “Evaluating the Evaluation of Cancer Driver Genes.” Proceedings of the National Academy of Sciences 113, no. 50 (December 13, 2016): 14330–35. https://doi.org/10.1073/pnas.1616440113. - 20/20+ machine learning method, ratiometric approach to predict cancer driver genes. Performance comparison of other methods, 20/20+, TUSON, OncodriveFML and MutsigCV are the top performers. https://github.com/KarchinLab/2020plus

Cancer gene signatures

  • Short-term TGFBR inhibition (mimicking decrease in TGFB pathway activity after pregnancy, p27+ progenitors) of prepubertal ACI inbred and Sprague Dawley outbred rats prevents estrogen- and carcinogen-induced breast cancer by likely increasing and depleting the pool of epithelial subpopulation of secretory basal cells (SBCs) with progenitor features. Bulk, scRNA-seq (10X Genomics, CellRanger, Seurat, monocle3, liana, more methods), SBC signature (336 genes, Supplementary Data 7), reanalysis of GSE106273 and GSE161529. GitHub, scRNA-seq data at GSE184095, R data object at DOI. Great Tweetorial by Simona Cristea. Polyak_2022_SBC_genes.xlsx
    Paper Alečković, Maša, Simona Cristea, Carlos R. Gil Del Alcazar, Pengze Yan, Lina Ding, Ethan D. Krop, Nicholas W. Harper, et al. “Breast Cancer Prevention by Short-Term Inhibition of TGFβ Signaling.” Nature Communications 13, no. 1 (December 7, 2022): 7558. https://doi.org/10.1038/s41467-022-35043-5.

Cancer driver mutations

For general variant interpretation databases, see SNP_notes/SNP annotations

  • Resources / databases for clinical interpretation of cancer variants, by Malachi Griffith, https://www.biostars.org/p/403117/

  • sigminer - an R package for SNP, CNV, DBS, InDel signature extraction from whole-exome data. NMF-based. Tested on tumor-notmal prostate cancer data. https://github.com/ShixiangWang/sigminer

    • Wang, Shixiang, Huimin Li, Minfang Song, Zaoke He, Tao Wu, Xuan Wang, Ziyu Tao, Kai Wu, and Xue-Song Liu. “Copy Number Signature Analyses in Prostate Cancer Reveal Distinct Etiologies and Clinical Outcomes.” Preprint. Genetic and Genomic Medicine, April 29, 2020. https://doi.org/10.1101/2020.04.27.20082404.

  • CANCERSIGN - identifies 3-mer and 5-mer mutational signatures, cluster samples by signatures. Based on Alexandrov method, Non-negative matrix factorization, explanation. Other tools - SomaticSignatures, SigneR, deconstructSigs, compared in Table 1. https://github.com/ictic-bioinformatics/CANCERSIGN

    • Bayati, Masroor, Hamid Reza Rabiee, Mehrdad Mehrbod, Fatemeh Vafaee, Diako Ebrahimi, Alistair Forrest, and Hamid Alinejad-Rokny. “CANCERSIGN: A User-Friendly and Robust Tool for Identification and Classification of Mutational Signatures and Patterns in Cancer Genomes.” BioRxiv, January 1, 2019, 424960. https://doi.org/10.1101/424960.

Cancer mutational signatures

  • MetaMutationalSigs - mutational signature analysis, aggregate and visualize results from different packages. Intro into mutational signatures, methods for signature analysis. Signature refitting packages, four packages (DeconstructSigs, MutationalPatterns, Sigfit, Sigminer). Input - VCF files. Output - heatmaps, csv tables.
    Paper Pandey, Palash, Sanjeevani Arora, and Gail L. Rosen. “MetaMutationalSigs: Comparison of Mutational Signature Refitting Results Made Easy.” Bioinformatics (Oxford, England), February 14, 2022, btac091. https://doi.org/10.1093/bioinformatics/btac091.

Databases

  • ENCODEC - integrative ENCODE resource for cancer genomics. Integrating multiple technologies (eCLIP, Hi-C, STARR-seq, other ENCODE data). Integrated annotation where genes are linked to concoding regulatory elements (Extended genes). Enables prioritization of key regulators, noncoding elements, variants associated with oncogenes. Downloadable annotation data, cell type-specific extended gene definition, annotated enhancer peaks, cell type-specific and merged networks.
    Paper Zhang, Jing, Donghoon Lee, Vineet Dhiman, Peng Jiang, Jie Xu, Patrick McGillivray, Hongbo Yang, et al. “An Integrative ENCODE Resource for Cancer Genomics.” Nature Communications 11, no. 1 (December 2020): 3696. https://doi.org/10.1038/s41467-020-14743-w.
  • Oncobox Atlas of Normal Tissue Expression (ANTE) - age-annotated RNA-seq from 142 solid tissue samples representing 20 organs from post-mortal human healthy donors of different age. Blood samples from 17 healthy volunteers. Integration with normal data with age from other databases (TCGA, ENCODE). Processed data at GSE120795 and FigShare.
    Paper Suntsova, Maria, Nurshat Gaifullin, Daria Allina, Alexey Reshetun, Xinmin Li, Larisa Mendeleeva, Vadim Surin, et al. “Atlas of RNA Sequencing Profiles for Normal Human Tissues.” Scientific Data 6, no. 1 (December 2019): 36. https://doi.org/10.1038/s41597-019-0043-4.

  • Refine.bio harmonizes petabytes of publicly available biological data into ready-to-use datasets for cancer researchers and AI/ML scientists. https://www.refine.bio/. Documentation, http://docs.refine.bio/en/latest/, GitHub, https://github.com/AlexsLemonade/refinebio.

  • Zehir, Ahmet, Ryma Benayed, Ronak H Shah, Aijazuddin Syed, Sumit Middha, Hyunjae R Kim, Preethi Srinivasan, et al. “Mutational Landscape of Metastatic Cancer Revealed from Prospective Clinical Sequencing of 10,000 Patients.” Nature Medicine 23, no. 6 (May 8, 2017): 703–13. https://doi.org/10.1038/nm.4333. - MSK-IMPACT study. Deep sequencing of 341-410 genes in 10,000 samples in multiple cancers. Focus on mutations, copy number alterations, fusions. Data at http://www.cbioportal.org/study?id=msk_impact_2017#summary, downloadable, includes clinical data for survival analysis.

  • Gendoo, Deena M.A., Michael Zon, Vandana Sandhu, Venkata Manem, Natchar Ratanasirigulchai, Gregory M. Chen, Levi Waldron, and Benjamin Haibe-Kains. “MetaGxData: Clinically Annotated Breast, Ovarian and Pancreatic Cancer Datasets and Their Use in Generating a Multi-Cancer Gene Signature,” November 12, 2018. https://doi.org/10.1101/052910. - MetaGxData package containing breast and ovarian cancer data, microarray- and RNA-seq gene expression and clinical annotations. Scripts to conduct genome-wide survival analysis for all genes. https://github.com/bhklab/MetaGxData

  • DepMap - Large-scale RNAi screen for cancer vulnerability genes in 501 cell lines from 20 cancers, shRNA silencing ~17,000 genes. DEMETER - Modeling and removal of shRNA off-target effects. 6 sigma cutoff of DEMETER scores to identify 769 differential gene dependencies. ATLANTIS model to predict other genes - MDPs, marker dependency pairs. Main data portal: https://depmap.org/portal/download/

    • Tsherniak, Aviad, Francisca Vazquez, Phil G. Montgomery, Barbara A. Weir, Gregory Kryukov, Glenn S. Cowley, Stanley Gill, et al. “Defining a Cancer Dependency Map.” Cell 170, no. 3 (July 2017): 564-576.e16. https://doi.org/10.1016/j.cell.2017.06.010. Supplemental tables, DepMap_TableS3_DependencyCorrelation.csv - Table S3. Gene Dependency-Dependency Correlations, pairs of genes essential for proliferation/viability. Columns: Gene symbol 1, Gene symbol 2, correlation (r), z_score. Source

  • CCLE2 data - CCLE characterization using sequencing technologies. Data described: RNA splicing, DNA methylation, Histone modification, miRNA expression, RPPA for 1072 cells. Data availability: https://portals.broadinstitute.org/ccle/data, https://depmap.org/portal/download/

    • Ghandi, Mahmoud, Franklin W. Huang, Judit Jané-Valbuena, Gregory V. Kryukov, Christopher C. Lo, E. Robert McDonald, Jordi Barretina, et al. “Next-Generation Characterization of the Cancer Cell Line Encyclopedia.” Nature, May 8, 2019. https://doi.org/10.1038/s41586-019-1186-3.

TCGA PanCancer

  • Catalog of cancer-associated gene alterations. PCAWG, ICGC and TCGA data analysis. Somatic alterations change gene expression, associated with splicing. Massive analysis results, figures, tables. All data links, All supplementary tables - S1 - characteristics of all samples, covariates. S5 - somatic eGenes, 649 eQTLs, coordinates, genes, p-values, effect sizes.
    Paper PCAWG Transcriptome Core Group, PCAWG Transcriptome Working Group, PCAWG Consortium, Claudia Calabrese, Natalie R. Davidson, Deniz Demircioğlu, Nuno A. Fonseca, et al. “Genomic Basis for RNA Alterations in Cancer.” Nature 578, no. 7793 (February 2020): 129–36. https://doi.org/10.1038/s41586-020-1970-0.

  • Pan-cancer analysis of somatic noncoding driver mutations. Raw data: https://docs.icgc.org/pcawg/data/, Processed data: https://dcc.icgc.org/releases/PCAWG/drivers, significantly recurring breakpoints and juxtapositions http://www.svscape.org/. Extended data figures and tables should be considered individually https://www.nature.com/articles/s41586-020-1965-x#additional-information

    • PCAWG Drivers and Functional Interpretation Working Group, PCAWG Structural Variation Working Group, PCAWG Consortium, Esther Rheinbay, Morten Muhlig Nielsen, Federico Abascal, Jeremiah A. Wala, et al. “Analyses of Non-Coding Somatic Drivers in 2,658 Cancer Whole Genomes.” Nature 578, no. 7793 (February 2020): 102–11. https://doi.org/10.1038/s41586-020-1965-x.

  • Pan-cancer study of metastatic solid tumour genomes, including whole-genome sequencing data for 2,520 pairs of tumour and normal tissue. Cancer-specific genomic variants, GATK/Strelka/Manta/MutationalPatterns/MSIseq. Online data and processed Supplementary tables with processed data. Table 4 - Recurring amplifications (a) and deletions (b) and associated target genes; Table 5 - Somatic driver catalogue; Table 6 - Germline driver catalogue; Table 7 - Gene Fusions; Table 9 - Actionable mutations. Priestley_2019_PanCancer data folder.

    Paper Priestley, Peter, Jonathan Baber, Martijn P. Lolkema, Neeltje Steeghs, Ewart de Bruijn, Charles Shale, Korneel Duyvesteyn, et al. “Pan-Cancer Whole-Genome Analyses of Metastatic Solid Tumours.” Nature, October 23, 2019. https://doi.org/10.1038/s41586-019-1689-y.

  • Alternative promoter activity in >18K RNA-seq samples of 42 cancer types (PCAWG, TCGA, GTeX). Tissue/cancer-specific deregulation, isoform diversity, variation in alternative promoters is associated with survival. H3K4me3 as a marker of active promoters. proActiv R package for estimation of promoter activity from RNA-seq data. Tweet

    • Demircioğlu, Deniz, Engin Cukuroglu, Martin Kindermans, Tannistha Nandi, Claudia Calabrese, Nuno A. Fonseca, André Kahles, et al. “A Pan-Cancer Transcriptome Analysis Reveals Pervasive Regulation through Alternative Promoters.” Cell, (September 2019)
    • Supplementary Material
    • Table S1 - The Transcript IDs with Corresponding Transcription Start Site IDs, Promoter IDs, and Gene IDs According to Gencode (Release 19) Annotations
    • Table S2 - The Transcription Start Site Coordinates for the Compiled Promoters (hg19)
    • Table S5 - The Complete List of Alternative Promoters, Including Tissue-Specific, Cancer-Associated, Multi-cancer-Associated, BRCA Molecular-Subtype-Associated, and Pan-Cancer-Associated Alternative Promoters
    • Table S7 - Promoters that Are Significantly Associated with Patient Survival, Related to Figures 5 and S5

  • Pan-Cancer atlas of alternative splicing events, called using SplAdder. Data in GFF3, HDF5, TXT formats: https://gdc.cancer.gov/about-data/publications/PanCanAtlas-Splicing-2018

    • Kahles, André, Kjong-Van Lehmann, Nora C. Toussaint, Matthias Hüser, Stefan G. Stark, Timo Sachsenberg, Oliver Stegle, et al. “Comprehensive Analysis of Alternative Splicing Across Tumors from 8,705 Patients.” Cancer Cell 34, no. 2 (August 2018): 211-224.e6. https://doi.org/10.1016/j.ccell.2018.07.001.

  • ATAC-seq data in 410 tumor samples from TCGA (23 cancer types). Correlation with gene expression predicts distal interactions. 18 clusters by cancer type. Data: hg19 coordinates of pan-cancer and BRCA-specific ATAC-seq peaks (Data S2), eQTLs (Data S5), peak-to-gene and enhancer-to-gene links (Data S7), and more https://gdc.cancer.gov/about-data/publications/ATACseq-AWG

    • Corces, M. Ryan, Jeffrey M. Granja, Shadi Shams, Bryan H. Louie, Jose A. Seoane, Wanding Zhou, Tiago C. Silva, et al. “The Chromatin Accessibility Landscape of Primary Human Cancers.” Edited by Rehan Akbani, Christopher C. Benz, Evan A. Boyle, Bradley M. Broom, Andrew D. Cherniack, Brian Craft, John A. Demchok, et al. Science 362, no. 6413 (2018). https://doi.org/10.1126/science.aav1898.

  • Papers and supplementary data from PanCancer publications. Clinical annotations, RNA-seq counts, RPPA, Methylation, miRNA, copy number, mutations in .maf format. https://gdc.cancer.gov/about-data/publications/pancanatlas

    • Ding, Li, Matthew H. Bailey, Eduard Porta-Pardo, Vesteinn Thorsson, Antonio Colaprico, Denis Bertrand, David L. Gibbs, et al. “Perspective on Oncogenic Processes at the End of the Beginning of Cancer Genomics.” Cell 173, no. 2 (April 5, 2018): 305-320.e10. https://doi.org/10.1016/j.cell.2018.03.033. - An overview of PanCancer Atlas.

  • The Pan-Cancer analysis by TCGA consortium, all papers. https://www.cell.com/pb-assets/consortium/pancanceratlas/pancani3/index.html

  • TCGA MC3 variant calling project. Eight variant callers. Protocols for filtering samples, variants. Public and controlled access MAF files at https://gdc.cancer.gov/about-data/publications/mc3-2017

    • Ellrott, Kyle, Matthew H. Bailey, Gordon Saksena, Kyle R. Covington, Cyriac Kandoth, Chip Stewart, Julian Hess, et al. “Scalable Open Science Approach for Mutation Calling of Tumor Exomes Using Multiple Genomic Pipelines.” Cell Systems 6, no. 3 (March 2018): 271-281.e7. https://doi.org/10.1016/j.cels.2018.03.002.

  • PCAGW - The PCAWG study is an international collaboration to identify common patterns of mutation in more than 2,800 cancer whole genomes from the International Cancer Genome Consortium. The project produced large amount data with many types including simple somatic mutations (SNVs, MNVs and small INDELs), large-scale somatic structural variations, copy number alterations, germline variations, RNA expression profiles, gene fusions, and phenotypic annotations etc. PCAWG data have been imported, processed and made available in the following four major online resources for download and exploration by the cancer researchers worldwide. http://docs.icgc.org/pcawg/

Pediatric

  • The Childhood Cancer Research Resources (CCRR) Portal maintained by Alex’s Lemonade Stand Foundation for Childhood Cancer. Projects as of September 2021: "Epigenomic profiling of neuroblastoma cell lines", "ATAC-Seq of neuroblastoma cell lines", "MYCN and MYC ChIP-Seq profiling in neuroblastoma cell lines", "Histone ChIP-Seq of neuroblastoma cell lines", "Transcriptomic Profiling of 39 Neuroblastoma Cell Lines", "A novel and highly effective mitochondrial uncoupling drug (MB1-47) in T-cell acute lymphoblastic leukemia", "A Tumor Suppressor Enhancer of PTEN in T-cell development and leukemia", "scRNAseq in Pten enhancer wild-type or deleted thymocytes, enriched for immature thymocyte stages (CD4-CD3-)", "ATAC-seq in NOTCH1-induced mouse T-ALLs", "Temporal, spatial, and genetic constraints contribute to the patterning and penetrance of murine Neurofibromatosis-1 optic glioma", "The Genomic Landscape of Juvenile Myelomonocytic Leukemia", "Mapping the Cellular Origin and Early Evolution of Leukemia in Down Syndrome"

  • St. Jude cloud Pediatric cancer resource. Whole genomes, exomes, transcriptomes, free download. Integrates other datasets. 135 subtypes of pediatric cancers, blood cancers, CNS and non-CNS solid tumors. hg38 processed data. Genomics platform for controlled access. PeCan - exploration of somatic variants. Visualization Community - integrated visualization of multi-omics and clinical information. ProteinPaint and Genome Paint. Data download

  • Treehouse - the Treehouse Childhood Cancer Initiative, a pediatric cancer-focused project at the University of California Santa Cruz Genomics Institute. Gene expression (counts, log2 TPM), deidentified clinical data. Compiles data from TCGA, TARGET, ICGC, St. Jude and other consortia, over 12K samples. Data curation challenges.

    Paper Learned, Katrina, Ann Durbin, Robert Currie, Ellen Towle Kephart, Holly C. Beale, Lauren M. Sanders, Jacob Pfeil, et al. “Barriers to Accessing Public Cancer Genomic Data.” Scientific Data 6, no. 1 (December 2019): 98. https://doi.org/10.1038/s41597-019-0096-4.

BRCA data

  • BCLncRDB - database of breast cancer-associated lncRNAs. 5,279 entries, including (I) Differentially expressed and methylated lncRNAs, (II) Stage and subtype-specific lncRNAs, and (III) Drugs, Subcellular localization, Sequence, and Chromosome information. Compared with non-BRCA-specific Lnc2cancer, LncRNADisease, LnCaNet, Lnc2catlas. Reprocessed data from GEO (GSE60689, GSE64790, GSE113851, and GSE119233), LncLocator, TCGA, literature. Download Excel files.
    Paper Kumar, Swapnil, Avantika Agarwal, and Vaibhav Vindal. “BCLncRDB: A Comprehensive Database of LncRNAs Associated with Breast Cancer.” Preprint. Bioinformatics, December 8, 2022. https://doi.org/10.1101/2022.12.05.519223.
  • A single-cell and spatially resolved atlas of human breast cancers. SCSubtype - intrinsic cell type subtyping for scRNA-seq data, to classify intratumor heterogeneity (ITTH) and derive seven gene modules (GMs), each containing 200 genes. Tested against PAM50 classification, pseudobulk method. Applied to 11 ER+, 5HER2+ and 10 TNBC breast primary tumors, over 130K cells, integrated using Seurat's anchoring method. inferCNV to separate normal/cancer cells. Immunophenotyping (CITE-seq), detailed immune milieu of breast cancer (lymphocytes and innate lymphoid cells, myeloid, stromal subclasses). Spatial transcriptomics to resolve heterogeneity. Nine clusters with unique cellular compositions and clinical outcomes. Supplementary information. Supplementary Table 4 - gene lists for each scSubtype (Basal_SC, Her2E_SC, LumA_SC and LumB_SC). Supplementary Table 5. - gene lists for the 7 ITTH gene-modules (GM1-7). Processed data on the Broad Institute Single Cell portal (text format), GSE176078. Raw data at EGAS00001005173. Spatial transcriptomics data at Zenodo.
    Paper Wu, Sunny Z., Ghamdan Al-Eryani, Daniel Lee Roden, Simon Junankar, Kate Harvey, Alma Andersson, Aatish Thennavan, et al. “A Single-Cell and Spatially Resolved Atlas of Human Breast Cancers.” Nature Genetics 53, no. 9 (September 2021): 1334–47. https://doi.org/10.1038/s41588-021-00911-1.
  • 37 PDX models from difficult to treat breast cancers. Whole-genome, RNA-seq, RPPA. PDXs conserve the molecular landscape and chemosensitivity of their corresponding patient tumors. Correlation of variant allele frequencies with primary tumors was highly variable but overall coding mutations are preserved. GSE142767 - RNA-seq data.
    Paper Savage, Paul, Alain Pacis, Hellen Kuasne, Leah Liu, Daniel Lai, Adrian Wan, Matthew Dankner, et al. “Chemogenomic Profiling of Breast Cancer Patient-Derived Xenografts Reveals Targetable Vulnerabilities for Difficult-to-Treat Tumors.” Communications Biology 3, no. 1 (December 2020): 310. https://doi.org/10.1038/s42003-020-1042-x.
  • BRCA GWAS identifies 32 novel susceptibility loci. 133,384 breast cancer cases and 113,789 controls, plus 18,908 BRCA1 mutation carriers (9,414 with breast cancer) of European ancestry, using both standard and novel methodologies that account for underlying tumor heterogeneity by estrogen receptor, progesterone receptor and human epidermal growth factor receptor 2 status and tumor grade. INQUISIT (Integrated eQTL and In Silico Prediction of GWAS Targets) to intersect candidate causal variants (CCVs) with functional annotation data from public databases to identify potential target genes. Scripts, TOP - R package for Two-stage polytomous logistic regression. Summary statistics at the Breast Cancer Association Consortium (BCAC) and at CIMBA – Consortium of Investigators of Modifiers of BRCA1/2. Supplementary information Table S5-S14 - variants identified by different methods in differenc conditions. Supplementary Table 15: INQUISIT analysis results, linking coordinates of variants with genes.
    Paper kConFab Investigators, ABCTB Investigators, EMBRACE Study, GEMO Study Collaborators, Haoyu Zhang, Thomas U. Ahearn, Julie Lecarpentier, et al. “Genome-Wide Association Study Identifies 32 Novel Breast Cancer Susceptibility Loci from Overall and Subtype-Specific Analyses.” Nature Genetics 52, no. 6 (June 2020): 572–81. https://doi.org/10.1038/s41588-020-0609-2

  • BREAST CANCER LANDSCAPE RESOURCE - A web portal to proteomics, transcriptomics, genomics and metabolomics of breast cancer.

    • Consortia Oslo Breast Cancer Research Consortium (OSBREAC), Henrik J. Johansson, Fabio Socciarelli, Nathaniel M. Vacanti, Mads H. Haugen, Yafeng Zhu, Ioannis Siavelis, et al. “Breast Cancer Quantitative Proteome and Proteogenomic Landscape.” Nature Communications 10, no. 1 (December 2019): 1600. https://doi.org/10.1038/s41467-019-09018-y. - Proteogenomics of breast cancer subtypes. ~10K proteins by LS-MS/MS. 9 samples for each of the five PAM50 subtypes. Protein expression partially recapitulates PAM50 subtypes, their own consensus clustering. High correlation with mRNA, less so for CNV. Correlation of 290 proteins that are FDA-approved drug targets. Online tool, supplementary Data 1 has the full protein expression matrix

  • BRCA GWAS in 122,977 European cases and 105,974 controls and 14,068 Asian cases and 13,104 controls. 65 new loci that are associated with overall breast cancer risk at p<5E-8.

    Paper NBCS Collaborators, ABCTB Investigators, ConFab/AOCS Investigators, Kyriaki Michailidou, Sara Lindström, Joe Dennis, Jonathan Beesley, et al. “Association Analysis Identifies 65 New Breast Cancer Risk Loci.” Nature 551, no. 7678 (November 2017): 92–94. https://doi.org/10.1038/nature24284
    Supplementary material https://www.nature.com/articles/nature24284#Sec23 - **Supplementary information** - Supplementary Tables 1, 2 and 24, Supplementary Table Guide. - **Supplementary Table 2** - Associations between previously reported breast cancer associated SNPs and breast cancer risk in the combined analysis of GWAS + iCOGS + OncoArray data for overall breast cancer. SNP rsIDs and genes. - **Supplementary Table 23** - BRCA SNPs, genes, categorized by pathways. - **Supplementary Table 24** - Pathway enrichment results and references. - **Supplementary data** - Supplementary Tables 3–5, 7, 9–12, 14–17, 21–23 and 25–33. - **Supplementary Tables 4-6** - SNP associations with ER+, ER-, Asian BRCA. - **Supplementary Table 7** - List of 65 breast cancer loci, with the number of credible risk variants and browser links. - **Supplementary Table 6** - 65 newly identified susceptibility loci for overall breast cancer. rsID, hg19 coordinates, genes. - **Supplementary Table 8** - Summary statistics for all variants for which the association with overall breast cancer in the combined dataset was significant at P<0.00001. SNPs, coordinates, p-values. - **Supplementary Table 13** - 2,221 credible variants at 65 novel loci, with genomic annotations, UCSC Genome Browser links. - **Supplementary Table 18** - gene-SNP eQTL associations significant at P<0.05 in the TCGA and METABRIC datasets. - **Supplementary Table 19/20** - Summary/Detailed INQUISIT gene prediction scores, separated to distal, promoter, coding, deteted with Oncoarray and published data.

  • Pooled shRNA screen of 77 breast cancer cell lines. siMEM algorithm to improve identification of susceptibility/driver genes. Known and novel genes. Processed gene expression and proteomics data, Supplementary data with BRCA genes

  • METABRIC paper. CNV/SNP and gene expression discovery and validation in 997 and 995 primary breast tumors, PAM50-classified. CNA segmentation using Circular Binary Segmentation (CBS). Integrative k-means clustering using 1000 genes cis-associated with CNVs, 10 clusters, characteristics of each. Affy/Illumina microarrays. Coordinates in hg18. Supplementary tables: Tables S2, S3 - clinical annotations; Tables S5, S12 - gene-centric Minimal common regions of alteration for CNAs (derived from CBS), stratified by ER status and PAM50 subtype for regions with frequency > 0.05; Tables S22, S23 - region-centric amplifications, S24 - deletions. Data on cBioPortal.

    Paper METABRIC Group, Christina Curtis, Sohrab P. Shah, Suet-Feung Chin, Gulisa Turashvili, Oscar M. Rueda, Mark J. Dunning, et al. “The Genomic and Transcriptomic Architecture of 2,000 Breast Tumours Reveals Novel Subgroups.” Nature 486, no. 7403 (June 2012): 346–52. https://doi.org/10.1038/nature10983.

PDX

  • Catalogs of PDX models: PDXfinder, EurOPDX, PRoXe, PDRM - Patient-derived Models Repository, BCM PDX portal, Jackson Laboratory PDX models.

  • PDXNet Image Repository, hosted on the Seven Bridges Cancer Genomics Cloud CGC - pan-cancer repository of nearly 1000 PDX and paired human progenitor H&E images, associated with genomic and transcriptomic data, clinical metadata, pathological annotations.

    Paper White, Brian S, Xing Yi Woo, Soner Koc, Todd Sheridan, Steven B Neuhauser, Shidan Wang, Yvonne A Evrard, et al. “A Pan-Cancer PDX Histology Image Repository with Genomic and Pathological Annotations for Deep Learning Analysis.” Preprint. Bioinformatics, October 27, 2022. https://doi.org/10.1101/2022.10.26.512745.
  • PDXnet portal - PDX models, data, workflow and discovery tools. The PDXNet Portal provides a way for researchers to learn about the PDX models, sequencing data (DNA and RNA), and PDX Minimum Information metadata tools generated by the network for public use. PDXnetwork funding opportunity.
    Paper Koc, Soner, Michael W Lloyd, Jeffrey W Grover, Nan Xiao, Sara Seepo, Sai Lakshmi Subramanian, Manisha Ray, et al. “PDXNet Portal: Patient-Derived Xenograft Model, Data, Workflow and Tool Discovery.” NAR Cancer 4, no. 2 (April 8, 2022): zcac014. https://doi.org/10.1093/narcan/zcac014.
  • Genomic characterization (WES: mutation, copy number, RNA-seq: fusion, transcriptomic profiles, and NCI-MATCH arms) of 536 PDX models across 25 cancer types. Mutations may disappear in PDXs. Multi-tool genomic variants calling (somaticwrapper, somatic.Mutect2_tumorOnly, germlinewrapper, CharGer, hatchet, msisensor2, all scripts on GitHub). Somatic mutations, copy number segment-level and gene-level, copy number chromosome arm-level, fusion, and gene expression data in text format at Figshare. Interactive viewer at PDX Variant Viewer at WUSTL.
    Paper Sun, Hua, Song Cao, R. Jay Mashl, Chia-Kuei Mo, Simone Zaccaria, Michael C. Wendl, Sherri R. Davies, et al. “Comprehensive Characterization of 536 Patient-Derived Xenograft Models Prioritizes Candidates for Targeted Treatment.” Nature Communications 12, no. 1 (December 2021): 5086. https://doi.org/10.1038/s41467-021-25177-3.
  • CNA analysis in 1,451 PDX and matched patient tumor (PT) samples from 509 PDX models. PT CNAs strongly conserved through passages. Intro about tumor evolution during engrafment and passaging, previous findings against (Ben-David 2017 reporting CNA divergence) and for CNA fidelity across passages. Five data types (SNPs, WES, WGS, RNA-seq, gene microarray), data agrees with each other except low accuracy for gene expression-based CNAs. PDX CNAs are present but not specifically enriched in cancer-related genes. PDX WES CNV (Xenome) Tumor-Normal Workflow and WES CNV Tumor-Normal Workflow ran on CGC Seven Bridges.
    Paper PDXNET Consortium, EurOPDX Consortium, Xing Yi Woo, Jessica Giordano, Anuj Srivastava, Zi-Ming Zhao, Michael W. Lloyd, et al. “Conservation of Copy Number Profiles during Engraftment and Passaging of Patient-Derived Cancer Xenografts.” Nature Genetics 53, no. 1 (January 2021): 86–99. https://doi.org/10.1038/s41588-020-00750-6.
  • Xeva (XEnograft Visualization & Analysis) - R/Bioconductor package for processing, visualization and integrative analysis of drug testing in PDX models. Follows the PDX minimum information (PDX-MI) standards. Handles replicate-based and 1x1x1 experimental design. XevaSet class to store pharmacological response (time vs. tumor volume) and metadata. The 'pdxModel' class to store PDX-MI variables. PDXE XevaSet object containing the Novartis PDX Encyclopedia (PDXE) data (3,470 unique PDX models tested across 57 treatments and derived from 191 patients spanning across 5 different cancer types). Functions to compute the association between genomic features and response to a drug (the slope of curves, angle between the mean control and treatment curves, tumor growth inhibition (TGI), area between the curves, linear mixed model, best average response (BAR), best response (BR), and mRECIST). The 'plotPDX' function displays tumor growth curves, plotting time versus tumor volume data. The 'waterfall' function visualizes population-level response for a given set of PDXs. 'plotmRECIST' displays mRECIST-based drug response as a heatmap, with drugs along heatmap rows and PDXs along columns. Gene expression (major oncomarkers) is consistent across passages. Genes in splicing and posttranscriptional mRNA processing pathways. Supplementary Data File 2 - Intra-class correlation coefficient (ICC) values for genes across all samples and stratified by tissue type, lower ICC = genes less stably expressed across passages. Supplementary Data File 5 - genomic feature (mutation, CNV, gene expression) associations with drug response, 21 drugs. Code to reproduce the paper, GitHub.
    Paper Mer, Arvind S., Wail Ba-Alawi, Petr Smirnov, Yi X. Wang, Ben Brew, Janosch Ortmann, Ming-Sound Tsao, David W. Cescon, Anna Goldenberg, and Benjamin Haibe-Kains. “Integrative Pharmacogenomics Analysis of Patient-Derived Xenografts.” Cancer Research 79, no. 17 (September 1, 2019): 4539–50. https://doi.org/10.1158/0008-5472.CAN-19-0349.
  • PDX-MI - minimal information standards about PDX models (clinical attributes, the process of implantation and passaging, quality assurance, etc., Table 1).
    Paper Meehan, Terrence F., Nathalie Conte, Theodore Goldstein, Giorgio Inghirami, Mark A. Murakami, Sebastian Brabetz, Zhiping Gu, et al. “PDX-MI: Minimal Information for Patient-Derived Tumor Xenograft Models.” Cancer Research 77, no. 21 (November 1, 2017): e62–66. https://doi.org/10.1158/0008-5472.CAN-17-0582.
  • Disambiguate - human-mouse read separation. Alignment to both genomes separately, selecting best matched genome. Supports multiple aligners, different strategies for STAR/BWA and TopHat/HiSat2.
    Paper Ahdesmäki, Miika J., Simon R. Gray, Justin H. Johnson, and Zhongwu Lai. “Disambiguate: An Open-Source Application for Disambiguating Two Species in next Generation Sequencing Data from Grafted Samples.” F1000Research 5 (2016): 2741. https://doi.org/10.12688/f1000research.10082.2.
  • 1 x 1 x 1 experimental design (PDX clinical trial or PCT) in vivo compound screens to assess the population responses to 62 treatments across six indications. Multi-omics correlation support compensates for single sample experiments. mRECIST criteria to measure drug response.
    Paper Gao, Hui, Joshua M. Korn, Stéphane Ferretti, John E. Monahan, Youzhen Wang, Mallika Singh, Chao Zhang, et al. “High-Throughput Screening Using Patient-Derived Tumor Xenografts to Predict Clinical Trial Drug Response.” Nature Medicine 21, no. 11 (November 2015): 1318–25. https://doi.org/10.1038/nm.3954.

Methylation

  • MEXPRESS - Gene-centric methylation and correlation with clinical parameters. http://mexpress.be/

  • Pancan-meQTL database of meQTLs across 23 TCGA cancer types. Cis-, trans-meQTLs, pancancer-meQTLs, survival meQTLs. SNP-, gene-, CpG-centric search for each cancer. Visualization, KM plots for survival. Download. http://bioinfo.life.hust.edu.cn/Pancan-meQTL/

    • Gong, Jing, Hao Wan, Shufang Mei, Hang Ruan, Zhao Zhang, Chunjie Liu, An-Yuan Guo, Lixia Diao, Xiaoping Miao, and Leng Han. “Pancan-MeQTL: A Database to Systematically Evaluate the Effects of Genetic Variants on Methylation in Human Cancer.” Nucleic Acids Research, September 7, 2018. https://doi.org/10.1093/nar/gky814.

  • Wanderer - An interactive viewer to explore DNA methylation and gene expression data in human cancer

Misc

  • AmpliconArchitect - reconstructs and classifies amplicons as circular or linear from WGS (short-read, paired-end). Walidated on simulated and experimental datasets, over a range of coverage (10X most optimal). A model in which focal amplifications arise due to the formation and replication of extrachromosomal DNA. Additional tools: AmpliconReconstructor, CycleViz.
    Paper Deshpande, Viraj, Jens Luebeck, Nam-Phuong D. Nguyen, Mehrdad Bakhtiari, Kristen M. Turner, Richard Schwab, Hannah Carter, Paul S. Mischel, and Vineet Bafna. “Exploring the Landscape of Focal Amplifications in Cancer Using AmpliconArchitect.” Nature Communications 10, no. 1 (December 2019): 392. https://doi.org/10.1038/s41467-018-08200-y.
Paper Wu, Sihan, Kristen M. Turner, Nam Nguyen, Ramya Raviram, Marcella Erb, Jennifer Santini, Jens Luebeck, et al. “Circular EcDNA Promotes Accessible Chromatin and High Oncogene Expression.” Nature, November 20, 2019. https://doi.org/10.1038/s41586-019-1763-5.
  • MSIseq - R package for microsatellite instability detection from exome mutation data (30X depth is OK). MSI is a form of hypermutation that occurs in some tumors due to defects in cellular DNA mismatch repair. MSI is characterized by frequent somatic mutations (i.e., cancer- specific mutations) that change the length of simple repeats (e.g., AAAAA...., GATAGATAGATA...) in tumor and matched normal DNA. Similar tools MSIsensor and mSINGS. Tested four classifiers (logistic regression, decision tree, random forest, naive Bayes from RWeka) on 10 variables (derived from number of mutations).
    Paper Ni Huang, Mi, John R. McPherson, Ioana Cutcutache, Bin Tean Teh, Patrick Tan, and Steven G. Rozen. “MSIseq: Software for Assessing Microsatellite Instability from Catalogs of Somatic Mutations.” Scientific Reports 5, no. 1 (October 2015): 13321. https://doi.org/10.1038/srep13321.