Molecular Systems Biology Peer Review Process File Sh3 Interactome Conserves General Function over Specific Form Transaction Report

(Note: With the exception of the correction of typographical or spelling errors that could be a source of ambiguity, letters and reports are not edited. The original formatting of letters and referee reports may not be reflected in this compilation.) Thank you again for submitting your work to Molecular Systems Biology. We have now heard back from two of the three referees who agreed to evaluate your manuscript, and we have decided to render a decision now to avoid further delay. As you will see from the reports below, the referees find the topic of your study of potential interest. They raise, however, substantial concerns, which, I am afraid to write, must preclude its publication in its present form. While the reviewers were cautiously positive about the goals of this work, the editor would like to emphasize that addressing their concerns appears to require additional experimental work (e.g. co-IP of AMPH-1/TBC-2 from live cells) and substantial new analyses. Reviewer #3 has important concerns with the evolutionary comparisons presented in this work, and feels that a more detailed analysis of binding specificity conservation is needed, as well as more cautious interpretations of the relative amounts of conservation between species pairs, given the low of number unambiguous orthologs in all comparisons. This reviewer also felt that a more thorough attempt should be made to place these findings in the context of other studies of evolution within interaction networks (the editor would also like to point a work by Habib et al, published very recently at our journal, The first reviewer also suggested that you consider incorporating some of the supplementary figures and data into the main manuscript to improve readability. Our formatting requirement would allow you to include an additional 2-3 figures and/or tables in the main manuscript. Please also deposit the new phospho-proteomic dataset in a public repository (e.g. PRIDE or PeptideAtlas), and include a confidential reviewer login with your revised manuscript.

yielded statistical descriptions of the relatedness of any two proteins, based on their domain compositions, and of the relationship between any two domains based on their co-occurrence among proteins. Complete linkage hierarchical clustering was then used to cluster rows and columns of the matrix and produce a two-way clustergram of the yeast and worm SH3 protein sets. The clustergrams were generated using the MATLAB Bioinformatics Toolbox.
Compared to yeast, worm has an expanded set of domains associated with the SH3 domain, including some domains specifically present in metazoa, such as L27 domains (LIN-2 and LIN-7) which hetero-oligomerize to assemble signaling and cell polarity complexes (Harris et al, 2002), PTB (Phosphotyrosine-Binding) domains, which organize signaling complexes involved in wide-ranging physiological processes including neural development, immunity, tissue homeostasis and cell growth (Uhlik et al, 2005), and NEBU (Nebulin repeat) domains, which function in actin cytoskeleton organization and regulation (Bjorklund et al;Nakagawa et al, 2009). These observations are consistent with the "domain accretion" model proposed by Koonin et al., where protein domain organization complexity increases with organism complexity (Koonin et al, 2000).

Thematic map of the worm SH3 endocytosis interactome (Supplementary Figure 6)
To analyze whether SH3-mediated PPIs are different from other PPIs in terms of connecting proteins within or between endocytosis-related functional modules, we employed thematic map analysis using two PPI networks (Supplementary Figure 6). The first network was a sub-network of the worm SH3 interactome containing all SH3 mediated PPIs among our expert-curated worm endocytosis protein list (Supplementary Table 10). The second network contains PPIs among our endocytosis proteins retrieved from iRefWeb Release 3.2 (Turner et al, 2010), not including any SH3 proteins. The two networks were merged in Cytoscape (Shannon et al, 2003) and visualized as a thematic map (Zhang et al, 2005) (using the Thematic Map Cytoscape plugin (Merico et al, 2011)), based on our functional annotation of worm endocytosis proteins (Supplementary Table   10). A thematic map is a simplified network view that highlights connections between modules (e.g., protein biological process annotations) within an original network. In this view, endocytosis protein annotations from the original network are represented as nodes that are connected if a PPI from the original network links proteins with the corresponding annotations.
P-values were computed by randomly shuffling the annotations of Supplementary Table 13 within the sub-networks.  PWMs, the highest similarity was used. Highlighted pairs correspond to domains on ortholog proteins (green) and on paralog proteins (red and black).

Supplementary Figure 2. Overlap between Y2H and phage display predicted PPIs for each
domain. Each worm SH3 domain with a phage-derived specificity profile represented as a PWM was used to score and rank all worm proteins for matches to this PWM. The plot shows the fraction of PPIs with a rank higher than the value on the x-axis for each domain (green curve).
The black dotted line indicates the expected distribution for random predictions.  (Breitkreutz et al, 2008). Edges represent interactions between worm proteins (blue), human orthologs (green) and conserved across worm and human (red). Diamonds indicate SH3-containing baits.

Supplementary Figure 4. Examples of rewiring occurring between worm and yeast SH3
interactomes. Case (i) corresponds to a conserved protein binding ligand but a different SH3 specificity. Case (ii) corresponds to a conserved SH3 specificity but a non-conserved binding ligand. Case (iii) corresponds to changes in both the SH3 specificity and the binding ligand. An example of each case is shown in the middle. All domains with the given rewiring case are shown at the right. Blue circles indicate worm proteins whose specificity or motif is not conserved. Green circles show yeast proteins whose specificity or motif is not conserved. Red circles show orthologs with conserved specificity or motifs.

Supplementary Tables
Supplementary Table 1. Worm SH3 domains used in phage display and yeast two-hybrid screens.