Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/047795

Tree inference for single-cell data

Katharina Jahn·Jack Kuipers·Niko Beerenwinkel
View original source ↗Export metadata ↓

Abstract

Understanding the mutational heterogeneity within tumours is a keystone for the development of efficient cancer therapies. Here, we present SCITE, a stochastic search algorithm to identify the evolutionary history of a tumour from noisy and incomplete mutation profiles of single cells. SCITE comprises a exible MCMC sampling scheme that allows the user to compute the maximum-likelihood mutation history, to sample from the posterior probability distribution, and to estimate the error rates of the underlying sequencing experiments. Evaluation on real cancer data and on simulation studies shows the scalability of SCITE to present-day single-cell sequencing data and improved reconstruction accuracy compared to existing approaches.

Source connections

Explore related subjects

Cancer BiologyPreprint

Keep this discovery

BibTeXRIS

Katharina Jahn, Jack Kuipers, Niko Beerenwinkel. 2016-04-09. Tree inference for single-cell data. https://doi.org/10.1101/047795

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2016

A molecular portrait of microsatellite instability across multiple cancers

Microsatellite instability (MSI) refers to the hypermutability of the cancer genome due to impaired DNA mismatch repair. Although MSI has been studied for decades, the large amount of sequencing data now available allows us to examine the molecular fingerprints of MSI in greater detail. Here, we analyze ~8000 exome and ~1000 whole-genome pairs across 23 cancer types. Our pan-cancer analysis reveals that the prevalence of MSI events is highly variable within and across tumor types including some in which MSI is not typically examined. We also identify genes in DNA repair and oncogenic pathways recurrently subject to MSI and uncover non-coding loci that frequently display MSI events. Finally, we propose an exomebased predictive model for the MSI phenotype that achieves high sensitivity and specificity. These results advance our understanding of the genomic drivers and consequences of MSI, and a comprehensive catalog of tumor-type specific MSI loci we have generated enables efficient panel-based MSI testing to identify patients who are likely to benefit from immunotherapy.

Cancer Biology↗
bioRxiv2016

Reduced CREB3L2 expression is associated with resistance to sorafenib and poor prognosis in ER-positive breast cancer

Sorafenib is a multikinase inhibitor that acts by inhibiting tumour growth and disrupting tumour microvasculature through anti-proliferative, anti-angiogenic, and pro-apoptotic effects. However, development of resistance to sorafenib often prevents its long-term efficacy. No validated biomarkers currently exist for appropriately selecting patients with cancer for sorafenib treatment. In the present study, we report that CREB3L2 expression in human breast cancer cell lines is a marker of response to sorafenib. Analysis of human breast cancer cell lines using Oncomine database and Genomics of Drug Sensitivity in Cancer (GDSC) database revealed an association between reduced expression of CREB3L2 and sensitivity to sorafenib. Wet lab experiment in five human breast cancer cell lines confirmed the association between reduced expression of CREB3L2 and sensitivity to sorafenib. Further, reduced expression of CREB3L2 was associated with poor Reccurence Free Survival in Luminal breast cancer. Our results suggest that CREB3L2 expression is a biomarker of response to sorafenib and outcome in breast cancer.

Cancer Biology↗
bioRxiv2016

A Hypothesis to Explain Cancers in Confined Colonies of Naked Mole Rats

Naked mole rats (NMRs) are subterranean eusocial mammals, known for their virtual absence of aging in their first 20 to 30 years of life, and their apparent resistance to cancer development. As such, this species has become an important biological model for investigating the physiological and molecular mechanisms behind cancer resistance. Two recent studies have discovered middle and late-aged worker (that is, non-breeding) NMRs in captive populations exhibiting neoplasms, consistent with cancer development, challenging the claim that NMRs are cancer resistant. These cases are possibly artefacts of inbreeding or certain rearing conditions in captivity, but they are also consistent with evolutionary theory.\n\nWe present field data showing that worker NMRs live on average for 1 to 2 years. This, together with considerable knowledge about the biology of this species, provides the basis for an evolutionary explanation for why debilitating cancers in NMRs should be rare in captive populations and absent in the wild. Whereas workers are important for maintaining tunnels, colony defence, brood care, and foraging, they are highly vulnerable to predation. However, surviving workers either replace dead breeders, or assume other less active functions whilst preparing for possible dispersal. These countervailing forces (selection resulting in aging due to early-life investments in worker function, and selection for breeder longevity) along with the fact that all breeders derive from the worker morph, can explain the low levels of cancer observed by these recent studies in captive colonies. Because workers in the field typically never reach ages where cancer becomes a risk to performance or mortality, those rare observations of neoplastic growth should be confined to the artificial environments where workers survive to ages rarely if ever occurring in the wild. Thus, we predict that the worker phenotype fortuitously benefits from anti-aging and cancer protection in captive populations.

Cancer Biology↗