Cell Ranger
Why a dot in the sample prefix hides FASTQs from Cell Ranger
Cell Ranger splits Illumina FASTQ names on a period. A dot inside the sample prefix makes --sample miss valid reads. How to see it, and how to repair the name without touching sequence.
What Cell Ranger is looking for
cellranger count finds FASTQs by prefix. The Illumina / 10x convention is [Sample]_S#_L###_R#_001.fastq.gz. The sample prefix is everything before _S. The rest of the tokens are separated by periods.
That last detail is the trap. Cell Ranger splits the filename on . to read the Illumina suffix. If the sample prefix itself contains a dot — Sample.A, donor.3, a library ID copied from a spreadsheet — the split no longer yields _S#_L###_R#_001. --sample then reports that the FASTQs are missing, even though the reads on disk are fine.
The repair
Replace dots (and other illegal characters) in the prefix with underscores, and leave the Illumina suffix intact. Sample.A_S1_L001_R1_001.fastq.gz becomes Sample_A_S1_L001_R1_001.fastq.gz. That is a syntax repair, not a change to scientific meaning. 10x documents the same filename pattern.
Do not rename blindly when the layout itself is inferred. NCBI SRA names such as SRR…_1.fastq.gz are a different convention; those should stay marked for review until you confirm read identity.
What the program will do
The planned Copper Scientific renamer will read filenames only. Sequence bytes are not opened, gzipped, or uploaded. It will print a plan, apply safe syntax repairs in the folder, and leave inferred layouts untouched until you confirm them.