@GenomeBiology: DeMixNB, a semi-reference-based deconvolution method for sparse count data. Tumor-cell transcript proportion is distinct from purity, and until now it's been challenging to estimate it in sparse-count data like miRNA-seq or Visium. https://t.co/EGVcEDVdhJ
Excited to share our latest research on bioRxiv 🧵
Current immunotherapy selection (TMB) excludes 85% of cancer patients. We found subclonal mutation load (sML) predicts better outcomes in low/moderate TMB patients.
Validated in prostate cancer trials: high sML predicted immunotherapy response where traditional criteria failed. sML captures tumor evolution info missing from current biomarkers.
We might be missing a whole population of "hidden responders" who could benefit from immunotherapy.
Beyond excited to share our lab's work in Nature Biotech today! It was a privilege working with members of @WenyiWang4 lab and other experts on this collaborative work! Congrats to @ShaolongCao@sxj307@JenniferWangMD
https://t.co/inkw9JUM1s
Very excited to share our paper in Nature Biotechnology today! Huge amount of work by @ShaolongCao@sxj307@JenniferWangMD et al! It is amazing to work with many experts across cancers. Every cancer tells its own story and our metric TmS can quantify it! https://t.co/7UiNTC745p
Very proud to release MuSE2.0 for somatic mutation calling, with ~50x speedup from MuSE1.0. It takes <10min to run WES data and 1hr to run WGS data from a tumor-normal pair. https://t.co/gp2dk1P0tR. Thanks to contributors: @mehrzadsamadi's team #Nvidia and Shuangxi Ji@sxj307!
We found a robust feature of cancer cell plasticity in bulk tumors using TmS, a mixed-genome-adjusted transcriptome deconvolution metric. This feature has demonstrated molecular and clinical relevance across cancers (brief summary below). Our preprint: https://t.co/VH5zIBseQQ.
We released CliP v1.2 today for fast subclonal reconstruction. With further optimization in computation, CliP finished running on ~9500 TCGA samples within 1 hour, and demonstrated accuracy within matched PCAWG samples. https://t.co/Qmu61n8C6c. Great job! @yujie_jiang679@sxj307
We released CliP v1.2 today for fast subclonal reconstruction. With further optimization in computation, CliP finished running on ~9500 TCGA samples within 1 hour, and demonstrated accuracy within matched PCAWG samples. https://t.co/Qmu61n8C6c. Great job! @yujie_jiang679@sxj307
How to make subclonal reconstruction lean and mean for whole genome sequencing data? We tried lasso-type penalty models and it works well!
See our bioRxiv paper: https://t.co/3nn7m01vc7 and the software tool: https://t.co/SDVOLeE60R.
“Using a very large number of samples, we observed that each tumor is different, but similar evolutionary patterns emerge that characterize cancer types,” says our Dr. Wenyi Wang: https://t.co/lylyp7iYOs @WenyiWang4#CancerMoonshot#EndCancer
Proud to present the final chapter of the PCAWG Evolution and Heterogeneity Working Group after 7 years: detailed pan-cancer characterisation of intra-tumour heterogeneity! #PCAWG https://t.co/Q1VFosuO9z
Scientifically sound article retracted after powerful company Herbalife and associates did not like the results and threatened to sue the journal.
This sounds awful. While journals do not want to retract papers with manipulated images. It is all about money, not science.
We are hiring a research scientist to work on anaplastic thyroid cancer. This is a great opportunity to analyze unique datasets and to understand tumor evolution with clinical impact, together with the wonderful thyroid cancer team at MDA. https://t.co/4EjyZF3H11. Please retweet!