*2.2. Secondary Structure of MLL43500–3630 and MLL44210–4280*

Disorder prediction profiles (Figure 1B,E) indicated that both protein regions have a significant disorder tendency. Disorder profile of MLL43500–3630 indicates a rather ambiguous disorder state, with prediction scores fluctuating around the 0.5 limit between ordered and disordered states. This disorder prediction might indicate a disordered region that has an elevated tendency to fold or a relatively unstable folded segment as well. Far-UV CD measurements revealed that MLL43500–3630 has a helical structure in isolation (Figure 1C). The CD spectrum of this region of MLL4 showed a typical alpha helical conformation with a pronounced double minimum at 208 and 220 nm. Secondary structure content calculation using the BeStSel algorithm [29,30] gave an α-helix content of ~36.2%, while another ~36% of the secondary structure content was characterized as "Others", which mainly corresponds to the disordered structure. Thermal unfolding of the observed helical structure was followed by gradually heating the sample to 100 ◦C while recording the absorbance at 220 nm (Figure 1C inset). The melting curve indicated a cooperative unfolding of the structure with

a melting point of 48 ◦C. The CD spectrum of the thermal denatured state is shown in Supplementary Figure S1, demonstrating a complete loss of structure at high temperatures.

MLL44210–4280 has a more pronounced disorder tendency, as demonstrated by the IUPred profile and is devoid of any predicted ANCHOR binding sites (Figure 1E). Its sequence contains a significant portion of glutamines (Figure 1D), but it does not contain Q stretches longer than 4 residues. Far-UV CD measurements confirmed the disorder predictions, indicating that the protein is mostly disordered in solution, with a considerable α-helical tendency. Secondary structure calculations gave a result of 16% α-helix and ~45% "Others" content, underlining that this segment of MLL4 is not fully disordered and contrary to interaction site predictions, might be involved in molecular recognition.
