|Supplementary Figure Legends
Supplementary Figure. 1. Consensus structures of domain 5. Consensus structures were made for each phylogenetic class using the introns listed below. RNAMotif descriptors were based on these structures and allowed the following deviations: 1) any nucleotide was allowed for positions not absolutely conserved (with the exception of the 2 nt bulge, which required the consensus sequences); 2) any segment that specified more than two unambiguous nucleotides (i.e., in the top strand of the 9 bp helix, the top strand of the 3-5 bp helix, the distal loop, the bottom strand of the 3-5 bp helix, and the bottom strand of the 9 bp helix) was allowed an additional nucleotide deviation from the consensus. For the ML class two descriptors were used, one described as above based only on bacteria (Panel G) and another based on a strict consensus of bacteria and organellar introns (Panel H). Introns used to make the consensus structures are found at (http://www.fp.ucalgary.ca/group2introns/) and are: (Panel A) B.a.I2, B.c.I2, B.c.I4, B.c.I5, B.me.I1, B.th.I1, Ba.sp.I1, C.d.I1, Cl.pe.I1, G.k.I1, E.f.I2, E.f.I3, E.f.I4, En.fm.I1, En.fm.I2, En.fm.I3; (Panel B) A.g.I1, A.v.I2, A.v.I4, B.f.I1, B.f.I2, B.f.I3, B.h.I1, B.sp.I1, B.t.I3, Bu.ce.I1, Bu.cp.I1, Bu.vi.I1, Bu.vi.I2, Bu.xe.I1, Bu.xe.I2, C.a.I1, D.h.I3, E.c.I6, E.c.I7, G.k.I2, G.s.I1, Ge.ur.I2, La.re.I1, M.sp.I1, N.e.I1, O.i.I1, P.a.I1, P.a.I2, P.ae.I1, P.p.l1, P.p.I3, P.r.I1, P.s.I1, Pe.ca.I2, Ph.p.I1, S.p.I1, S.t.I1, S.ag.I1, S.ma.I2, S.ma.I3, S.th.I1, So.us.I3, Sr.me.I3, Sy.fu.I1, Sy.wo.I2; (Panel C) B.j.I1, B.t.I4, E.a.I1, E.c.I1, E.c.I2, E.c.I3, Pa.de.I1, R.e.I1, Sr.t.I1, Sr.me.I1, Sh.dy.I1, UMB.I1; (Panel D) A.v.I5, B.j.I2, Cl.be.I3, P.l.I1, Ps.tu.I1, S.ma.I1; (Panel E) A.v.I1, A.v.I3, B.a.I1, B.c.I1, B.t.I1, B.me.I2, E.c.I5, Ha.ch.I1, M.a.I1-1, M.a.I1-2, M.a.I1-3, Ma.sp.I1, P.p.I2, P.ae.I2, Pe.lu.I1, Po.sp.I4, Ps.st.I1, Ps.st.I2, R.m.I1, R.r.I1, Sh.ba.I1, Sh.sp.I1, Th.e.I1, Th.e.I3, Th.e.I7, Th.e.I8, UA.I1, UA.I5, X.f.I1; (Panel F) An.v.I1, C.sp.I1, C.w.I1-1, C.w.I3, C.w.I5, C.w.I6, C.w.I7, G.v.I1, L.m.I1, N.p.I1, N.sp.I1, N.sp.I2, N.sp.I3, N.sp.I4, T.sp.I1, Tr.e.I1, Tr.e.I2, Tr.e.I3, Tr.e.I4, Tr.e.I5, Tr.e.I6, Tr.e.I7, Tr.e.I8, Tr.e.I9; (Panel G) B.c.I9, B.t.I2, Ba.fr.I1, D.h.I1, D.h.I2, D.h.I5, D.h.I7-1, L.l.I1, N.a.I1, N.a.I2, O.i.I2, OYP.I1, Sc.s.I1, St.py.I1; (Panel H) bacterial introns listed in panel G and the following organellar introns: Allomyces macrogynus cox1.I3, Marchantia polymorpha atpA.I1, Marchantia polymorpha cob1.I3, Marchantia polymorpha nad4.I1, Marchantia polymorpha rpl2.I1, Nicotiana tabacum rpl2.I2, Podospora anserina cox1.I1, Podospora anserina cox1.I4, Podospora anserina ND5.I4, Pylaiella littoralis cox1.I1, Pylaiella littoralis cox1.I2, Pylaiella littoralis cox1.I3, Saccharomyces cerevisiae cox1.I1, Saccharomyces cerevisiae cox1.I2, Schizosaccharomyces pombe cob1.I1, Schizosaccharomyces pombe cox1.I1, Schizosaccharomyces pombe cox2.I1, Venturia inaequalis cob1.I1.
Supplementary Figure 2. Domain V and VI secondary structures of likely truncated group II introns. These sequences were originally identified as candidate ORF-less introns, but appear to lack DI-DIV. GenBank accession numbers and coordinates are given in parentheses.
Supplementary Figure 3. Secondary structures of introns identified in the search for ORF-less introns.
A. Secondary structure of OYPI2 found in onion yellows phytoplasm (AP006628).
B. Secondary structure of O.i.I2 found in Oceanobacillus iheyensis HTE831 (NC_004193), which contains a highly degenerated ORF in domain IV (sequence not shown, indicated with solid line in DIV).