Simon Andrews did his first degree in Microbiology at the University of Warwick. After a brief period working for Sandoz pharmaceuticals he went on to do a PhD in protein engineering a the University of Newcastle with Harry Gilbert. During his PhD his interests moved from bench work toward the emerging field of bioinformatics, and he decided to follow this direction in his future career.
After completing his PhD Simon worked with the BBSRC IT Services where he developed and then presented a series of bioinformatics training courses in protein structure analysis to the BBSRC institutes. At one of these courses at Babraham he met John Coadwell who establised the Babraham Bioinformatics group and was then employed as the second member of the bioinformatics team. Since joining Babraham Simon has seen the group grow from two people to nine as the field has become far more prominent in the biological research community. He took over the running of the group in 2010.
The placental DNA methylation landscape is unique, with widespread partially methylated domains (PMDs). The placental "methylome" is conserved across mammals, a shared feature of many cancers, and extensively studied for links with pregnancy complications. Human trophoblast stem cells (hTSCs) offer exciting potential for functional studies to better understand this epigenetic feature; however, whether the hTSC epigenome recapitulates primary trophoblast remains unclear. We find that hTSCs exhibit an atypical methylome compared with trophectoderm and 1 trimester cytotrophoblast. Regardless of cell origin, oxygen levels, or culture conditions, hTSCs show localized DNA methylation within transcribed gene bodies and a complete loss of PMDs. Unlike early human trophoblasts, hTSCs display a notable absence of DNMT3L expression, which is necessary for PMD establishment in mouse trophoblasts. Remarkably, we demonstrate that ectopic expression of DNMT3L in hTSCs restores placental PMDs, supporting a conserved role for DNMT3L in de novo methylation in trophoblast development in human embryogenesis.
Promoters of developmental genes in embryonic stem cells (ESCs) are marked by histone H3 lysine 4 trimethylation (H3K4me3) and H3K27me3 in an asymmetric nucleosomal conformation, with each sister histone H3 carrying only one of the two marks. These bivalent domains are thought to poise genes for timely activation upon differentiation. Here, we show that asymmetric bivalent nucleosomes recruit repressive H3K27me3 binders but fail to enrich activating H3K4me3 binders, thereby promoting a poised state. Strikingly, the bivalent mark combination further promotes recruitment of specific chromatin proteins that are not recruited by each mark individually, including the lysine acetyltransferase (KAT) complex KAT6B. Knockout of KAT6B blocks neuronal differentiation, demonstrating that KAT6B is critical for proper bivalent gene expression during ESC differentiation. These findings reveal how readout of the bivalent histone marks directly promotes a poised state at developmental genes while highlighting how nucleosomal asymmetry is critical for histone mark readout and function.
Whole genome bisulfite sequencing (WGBS) has been the gold standard technique for base resolution analysis of DNA methylation for the last 15 years. It has been, however, associated with technical biases, which lead to overall overestimation of global and regional methylation values, and significant artifacts in extreme cytosine-rich DNA sequence contexts. Enzymatic conversion of cytosine is the newest approach, set to replace entirely the use of the damaging bisulfite conversion of DNA. The EM-seq technique utilizes TET2, T4-BGT, and APOBEC in a two-step conversion process, where the modified cytosines are first protected by oxidation and glucosylation, followed by deamination of all unmodified cytosines to uracil. As a result, EM-seq is degradation-free and bias-free, requires low DNA input, and produces high library yields with longer reads, little batch variation, less duplication, uniform genomic coverage, accurate methylation over a larger number of captured CpGs, and no sequence-specific artifacts.