{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/ca3b199628a9446f8175bfc1073cbf0b\" frameborder=\"0\" width=\"1280\" height=\"960\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":960,"width":1280,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":960,"thumbnail_width":1280,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/ca3b199628a9446f8175bfc1073cbf0b-9b840de7bee3c96d.gif","duration":250.773,"title":"Validating DIA single-cell proteomics identifications","description":"This Loom discusses how to use Diane for single-cell proteomics identifications with appropriate caution. The speaker evaluates the ReFentyl (turn-down) procedure on a single HeLa cell and finds it identifies more precursors but fewer proteins; ReFentyl-specific precursors have fewer spectrum ion matches on average (about 7 versus about 10 for common precursors), so confident peptide identification may require turning it off. For mouse liver single-cell data searched against a human plus mouse database, human-exclusive precursors exceed 6% across 14 cells, implying an actual “intrepid FDR” up to about 10% due to database size and likely software bias. The Loom also examines match-between-runs and notes low precursor overlap across single HeLa cells (28% appear in only one cell), with increased overlap from 0.54 to 0.74 and extra second-pass findings generally showing lower spectrum and ion support; users should report both first-pass and second-pass results."}