Iterative Fragmentation Improves the Detection of ChIP-seq Peaks for Inactive Histone Marks

SEAseq: a portable and cloud-based chromatin occupancy analysis suite

Background

Genome-wide protein-DNA binding is popularly assessed using specific antibody pulldown in Chromatin Immunoprecipitation Sequencing (ChIP-Seq) or Cleavage Under Targets and Release Using Nuclease (CUT&RUN) sequencing experiments. These technologies generate high-throughput sequencing data that necessitate the use of multiple sophisticated, computationally intensive genomic tools to make discoveries, but these genomic tools often have a high barrier to use because of computational resource constraints.

Results

We present a comprehensive, infrastructure-independent, computational pipeline called SEAseq, which leverages field-standard, open-source tools for processing and analyzing ChIP-Seq/CUT&RUN data. SEAseq performs extensive analyses from the raw output of the experiment, including alignment, peak calling, motif analysis, promoters and metagene coverage profiling, peak annotation distribution, clustered/stitched peaks (e.g. super-enhancer) identification, and multiple relevant quality assessment metrics, as well as automatic interfacing with data in GEO/SRA. SEAseq enables rapid and cost-effective resource for analysis of both new and publicly available datasets as demonstrated in our comparative case studies.

Conclusions

The easy-to-use and versatile design of SEAseq makes it a reliable and efficient resource for ensuring high quality analysis. Its cloud implementation enables a broad suite of analyses in environments with constrained computational resources. SEAseq is platform-independent and is aimed to be usable by everyone with or without programming skills. It is available on the cloud at https://platform.stjude.cloud/workflows/seaseq and can be locally installed from the repository at https://github.com/stjude/seaseq.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04588-z.

Genome-wide analysis of H3K4me3 and H3K27me3 modifications due to Lr28 for leaf rust resistance in bread wheat (Triticum aestivum)

Present study revealed a complex relationship among histone H3 methylation (examined using H3K4/K27me3 marks), cytosine DNA methylation and differential gene expression during Lr28 mediated leaf rust resistance in wheat. During the present study, genome-wide histone modifications were examined in a pair of near isogenic lines (NILs) (with and without Lr28 in the background of cv. HD2329). The two histone marks used included H3K4me3 (an activation mark) and H3K27me3 (a repression mark). The results were compared with levels of expression (using RNA-seq) and DNA methylation (MeDIP) data obtained using the same pair of NILs. Some of the salient features of the present study include the following: (i) large scale differential binding sites (DBS) were available for only H3K4me3 in the susceptible cultivar, but for both H3K4me3 and H3K27me3 in its resistant NIL; (ii) DBSs for H3K27me3 mark were more abundant (> 80%) in intergenic regions, whereas DBSs for H3K4me3 were distributed in all genomic regions including exons, introns, intergenic, TTS (transcription termination sites) and promoters; (iii) fourteen (14) genes associated with DBSs showed co-localization for both the marks; (iv) only a small fraction (7% for H3K4me3 and 12% for H3K27me3) of genes associated with DBSs matched with the levels of gene expression inferred from RNA-seq data; (v) validation studies using qRT-PCR were conducted on 26 selected representative genes; results for only 11 genes could be validated. The proteins encoded by important genes involved in promoting infection included domains generally carried by R gene proteins such as Mlo like protein, protein kinases and purple acid phosphatase. Similarly, proteins encoded by genes involved in resistance included those carrying domains for lectin kinase, R gene, aspartyl protease, etc. Overall, the results suggest a very complex network of downstream genes that are expressed during compatible and incompatible interactions; some of the genes identified during the present study may be used in future validation studies involving RNAi/overexpression approaches.