Fig. 3. LUMA structure prediction. (A) Kyte-Doolittle hydropathy plot. The question mark indicates part of the sequence that is predicted to form a TMD although the hydropathy profile does not clearly show a hydrophobic-sequence stretch. (B) Scheme of predicted domains in LUMA. Numbers indicate first and last amino acid of each predicted feature. Black and hatched boxes indicate putative TMDs. The grey box marks the PFAM domain DUF 1625. (C) Averaged secondary-structure prediction for LUMA according to analysis with the programs GOR1, NNPREDICT, HNN, SOPM and SOPMA. Grey, -helices; black, β-sheets (see also supplementary material Fig. S3). (D) Prediction of natively unfolded amino-acid stretches for LUMA. Grey, folded regions; black, natively unfolded regions. Numbers above indicate beginning and end of the large predicted natively unfolded region.