%0 Journal Article %J Cell Syst %D 2020 %T Community Assessment of the Predictability of Cancer Protein and Phosphoprotein Levels from Genomics and Transcriptomics. %A Yang, Mi %A Petralia, Francesca %A Li, Zhi %A Li, Hongyang %A Ma, Weiping %A Song, Xiaoyu %A Kim, Sunkyu %A Lee, Heewon %A Yu, Han %A Lee, Bora %A Bae, Seohui %A Heo, Eunji %A Kaczmarczyk, Jan %A Stępniak, Piotr %A Warchoł, Michał %A Yu, Thomas %A Calinawan, Anna P %A Boutros, Paul C %A Payne, Samuel H %A Reva, Boris %A Boja, Emily %A Rodriguez, Henry %A Stolovitzky, Gustavo %A Guan, Yuanfang %A Kang, Jaewoo %A Wang, Pei %A Fenyö, David %A Saez-Rodriguez, Julio %K Crowdsourcing %K Female %K Genomics %K Humans %K Machine Learning %K Male %K Neoplasms %K Phosphoproteins %K Proteins %K Proteomics %K Transcriptome %X

Cancer is driven by genomic alterations, but the processes causing this disease are largely performed by proteins. However, proteins are harder and more expensive to measure than genes and transcripts. To catalyze developments of methods to infer protein levels from other omics measurements, we leveraged crowdsourcing via the NCI-CPTAC DREAM proteogenomic challenge. We asked for methods to predict protein and phosphorylation levels from genomic and transcriptomic data in cancer patients. The best performance was achieved by an ensemble of models, including as predictors transcript level of the corresponding genes, interaction between genes, conservation across tumor types, and phosphosite proximity for phosphorylation prediction. Proteins from metabolic pathways and complexes were the best and worst predicted, respectively. The performance of even the best-performing model was modest, suggesting that many proteins are strongly regulated through translational control and degradation. Our results set a reference for the limitations of computational inference in proteogenomics. A record of this paper's transparent peer review process is included in the Supplemental Information.

%B Cell Syst %V 11 %P 186-195.e9 %8 2020 08 26 %G eng %N 2 %1 https://www.ncbi.nlm.nih.gov/pubmed/32710834?dopt=Abstract %R 10.1016/j.cels.2020.06.013 %0 Journal Article %J Nat Commun %D 2019 %T Community assessment to advance computational prediction of cancer drug combinations in a pharmacogenomic screen. %A Menden, Michael P %A Wang, Dennis %A Mason, Mike J %A Szalai, Bence %A Bulusu, Krishna C %A Guan, Yuanfang %A Yu, Thomas %A Kang, Jaewoo %A Jeon, Minji %A Wolfinger, Russ %A Nguyen, Tin %A Zaslavskiy, Mikhail %A Jang, In Sock %A Ghazoui, Zara %A Ahsen, Mehmet Eren %A Vogel, Robert %A Neto, Elias Chaibub %A Norman, Thea %A Tang, Eric K Y %A Garnett, Mathew J %A Veroli, Giovanni Y Di %A Fawell, Stephen %A Stolovitzky, Gustavo %A Guinney, Justin %A Dry, Jonathan R %A Saez-Rodriguez, Julio %K ADAM17 Protein %K Antineoplastic Combined Chemotherapy Protocols %K Benchmarking %K Biomarkers, Tumor %K Cell Line, Tumor %K Computational Biology %K Datasets as Topic %K Drug Antagonism %K Drug Resistance, Neoplasm %K Drug Synergism %K Genomics %K Humans %K Molecular Targeted Therapy %K mutation %K Neoplasms %K pharmacogenetics %K Phosphatidylinositol 3-Kinases %K Phosphoinositide-3 Kinase Inhibitors %K Treatment Outcome %X

The effectiveness of most cancer targeted therapies is short-lived. Tumors often develop resistance that might be overcome with drug combinations. However, the number of possible combinations is vast, necessitating data-driven approaches to find optimal patient-specific treatments. Here we report AstraZeneca's large drug combination dataset, consisting of 11,576 experiments from 910 combinations across 85 molecularly characterized cancer cell lines, and results of a DREAM Challenge to evaluate computational strategies for predicting synergistic drug pairs and biomarkers. 160 teams participated to provide a comprehensive methodological development and benchmarking. Winning methods incorporate prior knowledge of drug-target interactions. Synergy is predicted with an accuracy matching biological replicates for >60% of combinations. However, 20% of drug combinations are poorly predicted by all methods. Genomic rationale for synergy predictions are identified, including ADAM17 inhibitor antagonism when combined with PIK3CB/D inhibition contrasting to synergy when combined with other PI3K-pathway inhibitors in PIK3CA mutant cells.

%B Nat Commun %V 10 %P 2674 %8 2019 06 17 %G eng %N 1 %1 https://www.ncbi.nlm.nih.gov/pubmed/31209238?dopt=Abstract %R 10.1038/s41467-019-09799-2 %0 Journal Article %J Nature Communications %D 2018 %T A crowdsourced analysis to identify ab initio molecular signatures predictive of susceptibility to viral infection %A Fourati, Slim %A Talla, Aarthi %A Mahmoudian, Mehrad %A Burkhart, Joshua G. %A Klén, Riku %A Henao, Ricardo %A Yu, Thomas %A Aydın, Zafer %A Yeung, Ka Yee %A Ahsen, Mehmet Eren %A Almugbel, Reem %A Jahandideh, Samad %A Liang, Xiao %A Nordling, Torbjörn E. M. %A Shiga, Motoki %A Stanescu, Ana %A Vogel, Robert %A Pandey, Gaurav %A Chiu, Christopher %A McClain, Micah T. %A Woods, Christopher W. %A Ginsburg, Geoffrey S. %A Elo, Laura L. %A Tsalik, Ephraim L. %A Mangravite, Lara M. %A Sieberts, Solveig K. %B Nature Communications %V 9 %8 Jan-12-2018 %G eng %U http://www.nature.com/articles/s41467-018-06735-8http://www.nature.com/articles/s41467-018-06735-8.pdfhttp://www.nature.com/articles/s41467-018-06735-8.pdfhttp://www.nature.com/articles/s41467-018-06735-8 %N 1 %! Nat Commun %R 10.1038/s41467-018-06735-8