Data availability

All computer code, snoRNA search results, oligonucleotide primers, rRNA primer extensions gel, and other referenced data can be found on line [Lowe & Eddy, 1998]. All new snoRNAs (snR48, snR50-snR71) have been submitted to the Saccharomyces cerevisiae Genome Database (SGD; http://genome-www.stanford.edu/Saccharomyces/), and can be accessed directly by searching for SNR locus names (e.g., ``SNR50'', or ``SNR*''). Sequences are available in Genbank by accessions AF06461-AF064283 for snR48, snR50-snR71, respectively. Other yeast snoRNA Genbank accession numbers are as follows: snR190 and U14 (X96815), U18 (U12981), U24 (Z48760), snR13 (U16692), snR38 (U26012), snR39 (U26011), snR39b (X94605), snR40 (U26015), snR41 (U26016), snR47 (U56648), Z2-Z8 (Z69294-Z69300), Z9 (Z70300).

**Figure 4.1:** **Schematic diagram of snoRNA search algorithm.**
$\resizebox{!}{7.0in}{\includegraphics{figures/snoscan-algorithm.eps}}$

Table 4.1: Summary of states within snoRNA probabilistic model.

State numbers correspond to Figure 4.2. ``Ungapped HMM'' states represent fixed-length conserved sequence motifs. The state for the terminal stem is analogous, but models base pairs rather than single positions (e.g., a stochastic context-free grammar, SCFG [Durbin et al., 1998] instead of a hidden Markov model, HMM). Duration models for gaps are estimated from binned length distributions (e.g., the probability that a gap will be 11-20 nt, 21-30 nt, etc.). The guide state is a hidden Markov model dependent on the rRNA target sequence; it includes terms for the probability of starting the complementarity at a given position relative to rRNA (this probability is high near known methylation positions), the length of the complementarity, and the probability of mismatches and noncanonical base pairs in the complementarity. For each state, the most common feature (``consensus'') is shown to indicate the overall pattern we search for. The best, average, and worst feature scores are given for 41 methylation guide snoRNAs as an indication of the relative contribution of each state to the overall information in the model. For more detail, see the program source code [Lowe & Eddy, 1998].

State				Feature Score (bits)
number	[0pt]Feature	[0pt]Model	[0pt]Consensus	Best	Average	Worst
1	Terminal Stem	SCFG, 4-8 bp	6 bp (when present)	7.60	3.09	0.35
2	Box C	7 bp ungapped HMM	AUGAUGA	12.73	11.63	5.84
3	Gap	Duration model	Length 6-10 bp	-1.59	-2.09	-4.76
4	Guide Sequence	HMM	12 bp duplex	15.67	11.11	2.54
5	Box D'	4 bp ungapped HMM	CUGA	7.34	4.85	-3.74
6	Gap	Duration model	Length 36-45 bp	-1.59	-2.43	-5.36
7	Box D	4 bp ungapped HMM	CUGA	8.05	7.92	5.43
8	Gap	Duration model	Length 56-75 bp	-1.50	-2.10	-4.17
9	Guide Sequence	HMM	14 bp duplex	18.96	13.98	9.95

**Figure 4.2:** **Schematic of the probabilistic snoRNA model.**
States (boxes and ovals) are connected by transitions (arrows). Each numbered state is a probabilistic model of a sequence feature (Table 4.1). Transition probabilities are 1.0, except those shown for transitions 2 $\rightarrow$ 3 and 2 $\rightarrow$ 8, which account for the proportion of snoRNAs with a guide sequence adjacent to box D' and those with a guide sequence adjacent to box D, respectively.
$\resizebox{\textwidth}{!}{\includegraphics{figures/snoscan-model-v2.eps}}$

Table 4.2: C/D box snoRNAs in S. cerevisiae that function as methylation guides. (see legend previous page)

Target Predicted Verified Match/ Genomic Len Chr Position

[0pt]snoRNA Methyl Site Targ Site Targ Site Mism Placement

U14
SSU-Cm414 $\oplus$ (1) ND 13/1 cluster 5 123 10 139388 (C)

U18 LSU-Am647 $\oplus$ (1) $\oplus$ (3) 15/0 intronic 102 1 142357 (W)

LSU-Cm648 $\oplus$ (3)

U24 LSU-Cm1435 $\oplus$ (1) $\oplus$ (1) 14/0 intronic 87 13 500071 (C)

LSU-Am1447 $\oplus$ (1,2) $\oplus$ (1) 12/0

LSU-Gm1448

snR13
LSU-Am2278 $\bullet$ $\bullet$ 10/0 extragenic 107 4 1393187 (W)

LSU-Am2279 $\bullet$

snR38
LSU-Gm2812 $\oplus$ (1) ND 13/0 intronic 95 11 282830 (W)

snR39 LSU-Am805 $\oplus$ (1,2) $\oplus$ (3) 13/0 intronic 89 7 365249 (C)

snR39b LSU-Gm803 $\oplus$ (2) $\bullet$ 14/0 extragenic 95 7 366466 (C)

snR40 SSU-Gm1267 $\oplus$ (1,2) $\bullet$ 11/1 extragenic 97 14 89208 (W)

LSU-Um896 $\bullet$ $\bullet$ 12/0

snR41
SSU-Am541 $\bullet$ $\bullet$ 11/1 cluster 3 105 16 719237 (C)

SSU-Gm1123 $\oplus$ (1,2) $\bullet$ 20/1

snR47 SSU-Am619 $\bullet$ $\bullet$ 11/0 extragenic 99 4 541738 (C)

LSU-Am2218 $\oplus$ (3) $\bullet$ 12/0

snR48 LSU-Gm2788 $\bullet$ $\bullet$ 17/1 extragenic 112 7 609578 (W)

LSU-Gm2790 $\bullet$ $\bullet$ 15/0

snR50 (Z14) LSU-Gm865 $\bullet$ $\bullet$ 12/0 extragenic 89 15 259489 (W)

snR51 SSU-Am100 $\bullet$ $\bullet$ 16/1 cluster 3 107 16 718803 (C)

LSU-Um2726 $\bullet$ $\bullet$ 14/1

snR52 (Z13) SSU-Am420 $\bullet$ $\bullet$ 13/0 extragenic 92 5 431217 (C)

LSU-Um2918 $\bullet$ No 11/1

LSU-Gm2919 $\bullet$ No

snR53 SSU-Am796 $\bullet$ $\bullet$ 11/0 cluster 4 91 5 61699 (W)

snR54 SSU-Am973 $\bullet$ $\bullet$ 13/0 intronic 86 13 163620 (C)

snR55 (Z10) SSU-Um1265 $\bullet$ $\otimes$ 12/0 cluster 2 98 12 794793 (C)

snR56 SSU-Gm1425 $\bullet$ $\bullet$ 11/0 extragenic 86 2 88181 (W)

snR57 SSU-Gm1570 $\bullet$ $\bullet$ 15/0 cluster 2 88 12 795023 (C)

snR58 (Z12) LSU-Cm661 $\bullet$ $\bullet$ 13/0 extragenic 96 15 136182 (C)

snR59 LSU-Am805 $\bullet$ $\oplus$ (3) 14/0 intronic 78 16 173826 (W)

snR60 (Z15) LSU-Am815 $\bullet$ $\bullet$ 10/0 extragenic 104 10 348929 (C)

LSU-Gm906 $\bullet$ $\bullet$ 19/0

snR61 (Z11) LSU-Am1131 $\bullet$ $\bullet$ 11/0 cluster 2 90 12 794574 (C)

snR62 LSU-Um1886 $\bullet$ $\bullet$ 14/0 extragenic 100 15 409863 (C)

snR63 LSU-Am2254 $\bullet$ $\bullet$ 12/0 extragenic 255 4 323470 (C)

snR64 LSU-Cm2335 $\bullet$ $\bullet$ 11/0 extragenic 101 11 38812 (W)

snR65 LSU-Um2345 $\bullet$ $\bullet$ 11/0 extragenic 100 3 175909 (W)

snR66 (Z16) LSU-Um2415 $\bullet$ $\bullet$ 12/0 extragenic 86 14 586088 (W)

snR67 LSU-Gm2616 $\bullet$ $\bullet$ 11/2 cluster 4 82 5 61352 (W)

LSU-Um2721 $\bullet$ $\bullet$ 11/0

snR68 LSU-Am2637 $\bullet$ $\bullet$ 12/0 extragenic 136 9 97111 (W)

snR69 LSU-Cm2945 $\bullet$ $\bullet$ 18/3 extragenic 101 11 364418 (W)

snR70 SSU-Cm1637 $\bullet$ $\otimes$ 9/1 cluster 3 164 16 719047 (C)

snR71 LSU-Am2943 $\bullet$ $\bullet$ 9/1 extragenic 89 8 411228 (W)

snR72 (Z2) LSU-Am874 $\bullet$ $\bullet$ 14/0 cluster 1 91 13 298554 (W)

snR73 (Z3) LSU-Cm2956 $\bullet$ $\bullet$ 12/1 cluster 1 103 13 298306 (W)

snR74 (Z4) SSU-Am28 $\bullet$ $\bullet$ 13/0 cluster 1 80 13 298138 (W)

snR75 (Z5) LSU-Gm2286 $\bullet$ $\bullet$ 11/0 cluster 1 85 13 297915 (W)

snR76 (Z6) LSU-Cm2195 $\bullet$ $\bullet$ 13/1 cluster 1 101 13 297727 (W)

snR77 (Z7) SSU-Um578 $\bullet$ $\bullet$ 14/0 cluster 1 84 13 297506 (W)

snR78 (Z8) LSU-Um2419 $\bullet$ $\bullet$ 12/0 cluster 1 82 13 297277 (W)

snR79 (Z9) SSU-Cm1006 $\bullet$ $\bullet$ 16/1 extragenic 85 12 348511 (C)

$\begin{figure}% latex2html id marker 1882 \textbf{Legend for Table \ref{Table-ye... ...lo96}, (2) \cite{Nicoloso96}, (3) \cite{Ni98}; ND = Not determined. \end{figure}$

$\begin{figure}% latex2html id marker 1889 \textbf{Legend for Figures \ref{fig:ye... ...bly due to other types of neighboring nucleotide modifications. \par\end{figure}$

**Figure 4.3:** **Experimental confirmation of methylation guide function for snoRNAs snR60, snR50, snR72, and snR40.** Loss of 2'-O-methyl bands in low dNTP-concentration reactions for mutant strains (even lanes 8 and greater) relative to the wildtype strain (lane 6) indicates loss of the methylation site and thus functional confirmation. Polyacrylamide gel electrophoresis of primer extensions using ³²P end-labled primers annealing to 25S rRNA from position 914-939.
$\resizebox{!}{6.5in}{\includegraphics{figures/snoRNA-KO-gel1.eps}}$

**Figure 4.4:** **Experimental confirmation of methylation guide function for snoRNAs snR75, snR47, snR63, and snR13.** Loss of 2'-O-methyl bands in low dNTP-concentration reactions for mutant strains (even lanes 8 and greater) relative to the wildtype strain (lane 6) indicates loss of the methylation site and thus functional confirmation. Polyacrylamide gel electrophoresis of primer extensions using ³²P end-labled primers annealing to 25S rRNA from position 2305-2328.
$\resizebox{!}{6.5in}{\includegraphics{figures/snoRNA-KO-gel2.eps}}$

**Figure 4.5:** **Experimental confirmation of methylation guide function for snoRNAs snR40 and snR55** Loss of 2'-O-methyl band in low dNTP-concentration reaction for mutant strain (lanes 8 & 10) relative to the wildtype strain (lane 6) indicates loss of the methylation site and thus functional confirmation. Polyacrylamide gel electrophoresis of primer extensions using ³²P end-labled primers annealing to 18S rRNA from position 1291-1315.
$\resizebox{!}{6.5in}{\includegraphics{figures/snoRNA-KO-gel3.eps}}$

**Figure 4.6:** **Experimental confirmation of methylation guide function for snoRNA snR70.** Loss of 2'-O-methyl bands in low dNTP-concentration reactions for mutant strains (lane 8) relative to the wildtype strain (lane 6) indicates loss of the methylation site and thus functional confirmation. Polyacrylamide gel electrophoresis of primer extensions using ³²P end-labled primers annealing to 18S rRNA from position 1652-1675.
$\resizebox{!}{6.5in}{\includegraphics{figures/snoRNA-KO-gel4.eps}}$

**Figure 4.7:** **snoRNA primer extensions demonstrating expression of newly identified methylation guide snoRNAs.**
Reverse transcriptase primer extensions on total RNA from wildtype and snoRNA-disrupted strains. ³²P end-labeled primers complementary to internal snoRNA sequence for snR54 (lanes 1a,b), snR51 (lanes 2a,b), snR63 (lanes 3a,b), snR55 (lanes 4a,b), snR68 (lanes 5a,b), snR70 (lanes 6a,b), snR71 (lanes 7a,b) were used. snoRNA expression in wildtype RNA reactions (lanes 1a, 2a,...7a) was confirmed, as was loss of expression in snoRNA-deleted strains (lanes 1b, 2b,...7b). The snR54 internal snoRNA primer was included in all reactions as a positive control of intact RNA and active primer extension. RNA sequencing ladders of unrelated sequence are included on either side of snoRNA primer extensions for fragment size reference.
$\resizebox{!}{5.85in}{\includegraphics{figures/yeast-snoRT-gel.eps}}$

**Figure 4.8:** **Model for Addition of Adjacent 2'-O-methyls via the same snoRNA.** Listed are the four instances in yeast rRNA in which 2'-O-methyl groups occur just one nucleotide from other 2'-O-methyls. On the left hand side, base pairings between yeast rRNA and functionally confirmed (snR13, U24) or predicted (U18, snR52) methyl guide snoRNAs are depicted. Spacing between the D or D' box and rRNA sequence in each case presumably determines the location of 2'-O-methyl modification, invariably 5 bp from the end of the D/D' box [Kiss-Laszlo et al., 1996]). If the D' box is allowed to slide one nucleotide closer in via a single nucleotide bulge, the new placement of the D' box could conceivably guide addition of a second 2'-O-methyl group at the adjacent position. In each case, the single nucleotide bulge in the snoRNA would not necessarily disrupt required base pairings within the snoRNA/rRNA duplex.
$\resizebox{\textwidth}{!}{\includegraphics{figures/snoRNA-slide-meth.eps}}$