Defining the genes required for survival of Mycobacterium bovis in the bovine host offers novel insights into the genetic basis of survival of pathogenic mycobacteria

Supplementary dataset from "Defining the genes required for survival of Mycobacterium bovis in the bovine host offers novel insights into the genetic basis of survival of pathogenic mycobacteria"

Supplementary Figure legends

Figure S1. Illustration of the transposon insertions around the M. bovis genome. Sequencing of the input library showed that transposon insertions were evenly distributed around the genome and 27,419 of the permissible 66,931 thymine–adenine dinucleotide (TA) sites contained an insertion representing an insertion density of ~41%. The outer ring are the genomic coordinates, the blue lines represent transposon insertions and the gray boxes indicate regions of that did not have any insertions. Plot made with Circlize (Gu et al, 2014).

Figure S2. Diversity of the output library isolated from lung and thoracic lymph node lesions compared to the input library. On average, libraries recovered from lung lesions contained 14,456 unique mutants and those recovered from the lymph nodes contained an average of 16,210 unique mutants. Insertion density is represented as a proportion of the TA sites that contained insertions. The numbers on the x-axis refer to the sequencing file from that sample and come from individual animals (Bioproject ID: PRJNA816175, Submission ID: SUB11067380).

Figure S3. Volcano plots showing the distribution of log₂ fold-changes and -log₁₀ of adjusted p-values for representative lung (A) and lymph node (B) samples. Adjusted p-values (BH-fdr correction) < 0.000001 cluster at the limits of the plot and precision reflects the number of resampling iterations (10,000).

Figure S4. Scatterplot of mean log₂ fold change per gene for all lung samples against all thoracic lymph node samples. Correlation between mean log₂ fold change among genes between the tissues was calculated with Spearman's ranked correlation, = 0.878, p-value < 2.2e-16.

Figure S5. Fold-changes caused by transposon insertions in RD1^BCG and RD1^MIC in the lungs and lymph nodes of infected cattle. Boxplot for log₂ fold-changes in genes of the RD1^BCG region. Samples with adjusted p-values (BH-fdr corrected) <0.05 are indicated with purple points. Gene names highlighted in magenta have fewer than 5 TA sites located in the gene; too few to determine the statistical significance of changes in insertion levels with this method.

Supplementary Tables

Table S1. Sequencing statistics of the input and output transposon libraries. The numbers in the column labelled “filename” refers to the sequencing file from that sample and come from individual animals (Bioproject ID: PRJNA816175, Submission ID: SUB11067380).

Table S2. Tissues collected and scored for gross pathology. Tissues from head and neck lymph nodes (from the right and left sub-mandibular lymph nodes, the right and left medial retropharyngeal lymph nodes), thoracic lymph nodes (the right and left bronchial lymph nodes, the cranial tracheobronchial lymph nodes, the cranial and caudal mediastinal lymph nodes) and from lung lesions, were collected and scored.

Table S3. Log₂ fold-changes for insertions across the entire genome of M. bovis AF2122/97. Cells are coloured according to log₂ fold-change. Refer to the text for the gene groups in individual tabs.

Table S4. Custom transposon sequencing primers and adaptors used in sequencing of the transposon libraries.