Figure 3 from Mitochondrial tRNA-Derived Fragments as Candidate Metastasis-Modifying RNA

Le résumé fourni par la source

AlphaFold3 predictive modeling of mt-TR suggests different stability and D–T-loop structures among strains. A, Sequences for each mt-TR (from left to right: C, H, and F) modeled without posttranscriptional modifications (PTM) and displayed predicted instability/low confidence. Predicted template modeling “pTM” scores range from 0.26 to 0.28 on a scale of 0–1. B, Three rows of images are displayed with increased zoom from top to bottom. To improve confidence predictions and better reflect known principles of tRNA folding (37), sequences for each strain were modeled according to reported PTM of human mt-TR (m1A, m2G, and Ψ; ref. 31). PTM inclusion increased predicted stability and confidence scoring compared with that of unmodified sequences (pTM range, 0.75–0.77 vs. 0.26–0.28). The D–T-loop interactions have the lowest scoring confidence, which aligns with the commonly observed transcript end (Table 1) of fragments between C: U53/G54; H: U55/G56; F: U54/G55 (highlighted in green in the bottom two rows of images). Nucleotides and structure of predicted interactions between the D–T-loop differ across strains: T-loop G54 is predicted to interact with the D-loop in C mt-TR; T-loop U57 is predicted to interact with the D-loop in H mt-TR; both U54/G55 are predicted to interact specifically with A17 of the T-loop in C mt-TR. Specific predictive bonding and distances between the center of nucleotides are listed in tRF patterns in liver tissue. pIDDT, predicted local distance difference test.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.