Fig 1.
An overview of the study design and reanalysis pipeline.
QA: Quality assessment.
Table 1.
List of mouse and rat proteomics datasets that were reanalysed.
Fig 2.
(A) Number of canonical proteins identified across different mouse organs. The number within the parenthesis indicates the number of samples. (B) Range of normalised iBAQ protein abundances across different organs. The number within the parenthesis indicates the number of samples. (C) Canonical proteins identified across different datasets. The number within the parenthesis indicate the number of unique tissues in the dataset. (D) Range of normalised iBAQ protein abundances across different datasets. The number within parenthesis indicate the number of unique tissues in the dataset. (E) Comparison of total spectral data with the number of canonical proteins identified in each dataset and organ. (F) Distribution of canonical proteins identified across organs.
Fig 3.
(A) Number of canonical proteins identified across different rat organs. The number within the parenthesis indicates the number of samples. (B) Range of normalised iBAQ protein abundances across different organs. The number within the parenthesis indicates the number of samples. (C) Canonical proteins identified across different datasets. The number within the parenthesis indicate the number of unique tissues in the dataset. (D) Range of normalised iBAQ protein abundances across different datasets. The number within parenthesis indicate the number of unique tissues in the dataset. (E) Comparison of total spectral data with the number of canonical proteins identified in each dataset and organ. (F) Distribution of canonical proteins identified across organs.
Fig 4.
(A) Heatmap of pairwise Pearson correlation coefficients across all mouse samples. The colour represents the correlation coefficient and was calculated using the bin transformed iBAQ values. The samples were hierarchically clustered on columns and rows using Euclidean distances. (B) PCA of all samples, using the binned protein abundances as input, coloured by the organ types. (C) PCA of all samples coloured by their respective dataset identifiers. The numbers in parenthesis indicate the number of datasets for each organ. Binned values of canonical proteins quantified in at least 50% of the samples were used to perform the PCA.
Fig 5.
(A) Heatmap of pairwise Pearson correlation coefficients across all rat samples. The colour represents the correlation coefficient and was calculated using the bin transformed iBAQ values. The samples were hierarchically clustered on columns and rows using Euclidean distances. (B) PCA of all samples coloured by the organ types. (C) PCA of all samples coloured by their respective dataset identifiers. The numbers in parenthesis indicate the number of datasets for each organ. Binned values of canonical proteins quantified in at least 50% of the samples were used to perform the PCA.
Fig 6.
Organ specificity of canonical proteins in (A) mouse and (B) rat.
Table 2.
Analysis of the top three GO terms for each organ in mouse and rat using the elevated organ-specific and group-specific canonical proteins as described in the ‘Methods’ section.
Table 3.
Homologs identified in mouse and rat datasets when compared with the background list of genes (corresponding to canonical proteins) identified in human datasets (Supplementary File 2 in [14]).
Fig 7.
Comparison of protein abundances (in ppb) between one-to-one mapped orthologs of mouse, rat and human in various organs.
(A) Pairwise correlation using normalised protein abundances of human and mouse orthologues. (B) Human and rat orthologs. (C) Mouse and rat orthologs. (D) As an example, the comparisons of binned protein expression of ten randomly sampled orthologs are shown. Data corresponding to all cases (as reported in panel D) are available in S7 File and the corresponding illustration of binned values is available in S8 File. Orthologs in (D) are shown using their human gene symbol.
Fig 8.
Visualisations generated using the UMAP algorithm to show the relationships between human, mouse, and rat samples.
(A) Shows the relationship of all samples, particularly showing strong relationship between biological systems. (B) Shows the protein abundancy of 3 example gene orthologs (SH3GL2, MYOZ2 and PYROXD2), within each sample. Human baseline protein expression data was generated in [14].
Fig 9.
Pathway analysis performed using the canonical proteins, showing the statistically significant representative pathways (p-value < 0.05) in (A) mouse and (B) rat organs.