The start of a new year is always a nice time to look back and take stock of the past year, and look forward and set some goals for the coming year. I spent the entirety of 2018 as an AgreenSkills+ Visiting Scholar at UWM in Milwaukee, Wisconsin, which has been (and continues to be!) a very rich experience that has given me the chance to broaden my understanding of statistical genetics and genomics and expand my skill set.
I participated in several projects that were broadly focused on the integrative analysis of multi-omics data. In particular, we submitted and eventually published our work on the decomposition of gene expression variation into parts attributable to copy number alterations, methylation, somatic mutations, transcription factor expression, microRNA expression, and germline genetic variation in cancer genomes. In a collaboration with Dr. Ava Udvadia at UWM, we also worked on a pretty cool short time-course analysis of matched RNA-seq and ATAC-seq data in a study on optic nerve regeneration in zebrafish; we hope to have a manuscript submitted in the early weeks of 2019. We submitted our methodological work (and associated R package maskmeans) on multi-view cluster splitting and aggregation, which was illustrated using multi-omic (RNA-seq, miRNA-seq, methylation) breast cancer data. Towards the end of the year, we started work on a new project involving individualized pathway-specific analyses; tune in soon for more details on that. Thanks to my collaboration with Paul Auer, I also learned a lot about the mechanics of GWAS and Mendelian randomization analyses, in particular for the large-scale UK Biobank and TOPMed resources.
This was also a year where several recent collaborations all seemed to come to fruition at once, including 4 that were published, 1 that was accepted for publication, and 3 that were submitted.
I gained a lot of new knowledge about R. In particular, thanks in part to prep work for the Data Management and Visualization class I taught at UWM last spring, I finally got a good handle on the tidyverse (dplyr, tidyr), web scraping basics (rvest), creating and extracting information from SQLite databases in R (RSQLite), and creating Shiny apps. I used my new Shiny coding skills to develop the EDGE in TCGA app, as well as 2 internal apps with collaborators that we hope to publish early this year. In addition, I finally got around to vastly improving my website using blogdown.
I didn’t travel nearly as much this year as I have in the recent past (but that was partially compensated by lots and LOTS of video conference calls). This wasn’t entirely accidental, as I wanted to take the time to settle into life in Milwaukee after the very stressful move with my family from France in late 2017. I had a first trip out to San Diego in January for PAG XXVI, where I presented a talk on our coseq Bioconductor package as well as work done for the Fr-AgENCODE project. My main work trip this year was back to France in June to attend the final meeting of the MixStatSeq ANR grant led by my colleague Cathy Maugis-Rabusseau, which also gave me the opportunity to reconnect with some of my INRA colleagues. For personal travel, I got to see family and friends all over the country (Wisconsin, Minnesota, Illinois, Kansas, Maine, North Dakota, Pennsylvania, Michigan) and take a family camping trip in Southern California.