Seeking lost proteome: translated small ORFs and bioactive microproteins in non-Hodgkin lymphoma

The human genome is believed to contain approximately 20,000 protein-coding genes, originally annotated based on criteria for predicting synthesis of stable proteins, including an arbitrary size threshold of 100 amino acids. Technological advances such as Ribosome footprinting (RiboSeq), which reveals the position of every translating ribosome, have abruptly challenged these assumptions, revealing widespread translation of thousands of previously unannotated, non-canonical, small open reading frames (smORFs) located in “untranslated” regions or “non-coding” RNAs. The biological impact of non-canonical translation is beginning to emerge, revealing smORFs that encode novel microproteins with potent biological functions. However, the vast majority of smORFs remain completely uncharacterised at a functional level. This suggests the potential for an entire layer of bioactive molecules that remains almost entirely unexplored. Initial studies were constrained by limited depth of RiboSeq data (often from a single tissue type or cell line), variation in the ORF-calling pipelines used, and the imperfections of current experimental technologies to screen bioactive microproteins. These constraints are evident from the minimal overlap in smORFs identified across studies, the diverse estimates of predicted smORF numbers, and the comparatively low number of microproteins to which biological function has been attributed. Our collaborative study will leverage the largest collection of pooled RiboSeq data ever analysed, spanning multiple tissue types. Our preliminary findings have already revealed thousands of smORFs, many with evidence of stable expression and apparent cell-essential function. We will deploy optimised and standardised analytical pipelines that incorporate innovative approaches such as translation factor binding data and tissue specific isoform usage to improve smORF calling precision. To overcome CRIPSR-associated limitations, we will exploit our optimised protocols for base-editor mutagenesis, generating a comprehensive base editor library targeting the start codon of every predicted smORF in the human translatome. We will perform start codon mutagenesis screens using cell lines and primary cells. These data will be integrated with ultrasensitive, deep proteomic profiling, single cell perturbational transcriptomics and co-expression network analysis to reveal stably-expressed microproteins, and resolve molecular functions for hundreds of bioactive smORFs. We will establish disease relevance of smORFs by integrating with variant data from thousands of cancer whole genomes and from 0.5million UK Biobank participants. We will develop deep mechanistic understanding of the function of selected microproteins to identify their contribution to cell biology and disease and to identify exploitable therapeutic vulnerabilities. Our computational analysis will encompass a broad, multi-tissue platform. However, our mechanistic investigation will focus onto non-Hodgkin lymphoma (NHL), where we can leverage strong RiboSeq data and the most advanced, already-optimised model systems for functional genomic screening. NHL is the 5th commonest human cancer and a significant cause of global morbidity and mortality. The slow progress in developing effective new therapies suggests gaps in our understanding of lymphoma biology. Whilst previous research has predominantly focused on the classical protein-coding genome, we propose that crucial insights into NHL biology may lie within the unexplored, non-canonical proteome. Ultimately, our study will provide a comprehensive functional compendium of non-canonical translation across the human genome, elucidating novel mechanisms that will reshape future research direction in fundamental cell biology and lead to novel therapeutic approaches in NHL. The computational and experimental tools developed through this study, including pan-start codon base editor library, will prove valuable resources for researchers as this emerging field of non-canonical translation continues to expand rapidly.