Tumors are complex tissues of cancerous cells surrounded by a heterogeneous cellular microenvironment with which they interact. Single-cell sequencing enables molecular characterization of single cells within the tumor. However, cell annotation—the assignment of cell type or cell state to each sequenced cell—is a challenge, especially identifying tumor cells within single-cell or spatial sequencing experiments. Here, we propose ikarus, a machine learning pipeline aimed at distinguishing tumor cells from normal cells at the single-cell level. We test ikarus on multiple single-cell datasets, showing that it achieves high sensitivity and specificity in multiple experimental contexts.

No abstract

The use of RNA sequencing from wastewater samples is proven to be a valuable way for estimating infection dynamics and circulating lineages of SARS-CoV-2. This approach has the advantage of being independent from patient population testing and symptomatic disease courses. However, it is equally important to develop easily accessible and scalable tools which can highlight critical changes in infection rates and dynamics over time across different locations given the sequencing data from the wastewater. Here we provide the first analysis of variant dynamics in Germany using wastewater sequencing and present PiGx SARS-CoV-2, a bit-by-bit reproducible end-to-end pipeline with comprehensive reports. To our knowledge, this is the first pipeline that includes all steps from raw-data to shareable reports, additional taxonomic analysis, deconvolution and geospatial time series analysis. Using our pipeline on a dataset of wastewater samples, from different locations across Berlin, over the time period from February 2021 to June 2021, we could reconstruct the dynamic of the Variant of Concern (VoC) B.1.1.7 (alpha). Additionally, we detected the unique signature mutation M:T26767C for the VoC B.1.617.2 (delta) and its raise in early June. We also show that SARS-CoV-2 mutation load measured from wastewater sequencing is correlated with actual case numbers and it has potential to be used in a predictive manner. All in all, our study provides additional evidence that systematic wastewater analysis using sequencing and computational methods can be used for modeling the infection dynamics of SARS-CoV-2. In addition, the results show that our tool can be used to tease out new mutations and to detect any emerging new lineages of concern before clinical detection. Our approach can support efforts for establishing continuous monitoring and early-warning projects for COVID-19 or any other infectious disease.

Tumors are highly complex tissues composed of cancerous cells, surrounded by a heterogeneous cellular microenvironment. Tumor response to treatments is governed by an interaction of cancer cell intrinsic factors with external influences of the tumor microenvironment. Disentangling the heterogeneity within a tumor is a crucial step in developing and utilization of effective cancer therapies. Single cell sequencing has the potential to revolutionize personalized medicine. In cancer therapy it enables an effective characterization of the complete heterogeneity within the tumor. A governing challenge in cancer single cell analysis is cell annotation, the assignment of a particular cell type or a cell state to each sequenced cell. We propose Ikarus, a machine learning pipeline aimed at solving a perceived simple problem, distinguishing tumor cells from normal cells at the single cell level. Automatic characterization of tumor cells is a critical limiting step for a multitude of research, clinical, and commercial applications. Automatic characterization of tumor cells would expedite neoantigen prediction, automatic characterization of tumor cell states, it would greatly facilitate cancer biomarker discovery. Such a tool can be used for automatic annotation of histopathological data, profiled using multichannel immunofluorescence or spatial sequencing. We have tested ikarus on multiple single cell datasets to ascertain that it achieves high sensitivity and specificity in multiple experimental contexts.