Analysis of chromatin organization and gene expression in T cells identifis functional genes for rheumatoid arthritis

scripts to accompany the paper "Analysis of chromatin organization and gene expression in T cells identifies functional genes for rheumatoid arthritis" authored by Jing Yang, Amanda McGovern, Paul Martin, Kate Duffus, Xiangyu Ge, Peyman Zarrineh, Andrew P Morris, Antony Adamson, Peter Fraser, Magnus Rattray & Stephen Eyre. Scripts are based on R and presented in Jupyter notebook.

Table of contents:

• ATACseq
• RNAseq
• CHiC
• HiC
• Linking_CHiC_ATACseq_RNAseq

ATACseq

• Heatmap: heatmap of ATAC-seq data based on the clustering results from ATACseq_clustering_Fig_3c.ipynb.
• Clustering: hierarchical Gaussian Processing (GP) clustering of dynamic ATAC-seq time course data.
• QC: quality check of ATAC-seq data.
• Dynamics: calculate Loglikelihood ratio (LR) and bayesian information criteria (BIC) for ATAC-seq data.

RNAseq

• Correlation between exons and introns: illustrate the correlations between exons and introns counts data from RNA-seq samples. Figures are shown in Supplementary Fig. 1a-c.
• PCA: display PCA results of RNA-seq data. Figure is shown in Supplementary Fig. 1d
• QC: compare the gene expression data from this study with the data from "Ye, C. J., et al. "Intersection of population variation and autoimmunity genetics in human T cell activation." Science 345.6202 (2014): 1254665". Figures are shown in Supplementary Fig. 1e-i

CHiC

• QC: compare CHi-C interactions with similar data from "Burren, O.S. et al, Chromosome contacts in activated T cells identify autoimmune disease candidate genes, Genome Biology, 18 (165) (2017)" as shown in Supplementary Fig. 6a.
• Clustering: clustering of CHi-C data and generate Supplementary Fig. 6b.

HiC

• Interactions: generate the interaction matrices for Hi-C data at different time. Upper triangular part of Hi-C data of chr1 is shown in upper panel of Fig. 2a
• QC: compute the Stratum adjusted Correlation Coefficient (SCC) between Hi-C datasets (Fig. 2b) and A/B compartments (Fig. 2c), respectively.
• SNPs: illustrate the overlap of rheumatoid arthritis SNPs and A/B compartments at different times as shown in the lower panel of Fig. 2a.
• Dynamics: illustrate the percentage changes of TADs, compartment As and compartment Bs over time as illustrated in Supplementary Fig. 7

Linking_CHiC_ATACseq_RNAseq

Directory for producing correlation density plots between linked CHi-C, ATAC-seq and RNA-seq data.

• Correlations: generate the correlation density maps of linked CHi-C, ATAC-seq and RNA-seq data under varied distance ranges around promoters as shown in Fig. 4a.
• Dynamics: dynamics of CHi-C, ATAC-seq, RNA-seq at different levels.
  • Dynamics in foldchange: display the log2 fold change of ATAC-seq vs gene and CHi-C vs gene, as shown in Fig. 4b-c
  • Dynamics in original data: display dynamic patterns of CHi-C and ATAC-seq associated with different gene clusters.
  • Dynamics at CTCF/H3K27AC binding sites: display dynamics of CHi-C, ATAC-seq and Gene expression linked to ATAC-seq peaks intersected with CTCF/H3K27AC binding sites.
  • Dynamics at transcription factor binding sites: display dynamics of CHi-C at ATAC-seq peaks intersected with CTCF/H3K27AC/NFKB binding sites.

For more clarification, please feel free to contact Jing Yang via : Jing.Yang@manchester.ac.uk