The sequence.index file is a tab delimited file containing all the meta data you should need to download and subset the files on this ftp site by individual, library, experiment and sequencing technology. The columns are 1. FASTQ_FILE, path to fastq file on ftp site 2. MD5, md5sum of file 3. RUN_ID, SRA/ERA run accession 4. STUDY_ID, SRA/ERA study accession 5. STUDY_NAME, Name of stury 6. CENTER_NAME, Submission centre name 7. SUBMISSION_ID, SRA/ERA submission accession 8. SUBMISSION_DATE, Date sequence submitted, YYYY-MM-DAY 9. SAMPLE_ID, SRA/ERA sample accession 10. SAMPLE_NAME, Sample name 11. POPULATION, Sample population 12. EXPERIMENT_ID, Experiment accession 13. INSTRUMENT_PLATFORM, Type of sequencing machine 14. INSTRUMENT_MODEL, Model of sequencing machine 15. LIBRARY_NAME, Library name 16. RUN_NAME, Name of machine run 17. RUN_BLOCK_NAME, Name of machine run sector 18. INSERT_SIZE, Submitter specifed insert size 19. LIBRARY_LAYOUT, Library layout, this can be either PAIRED or SINGLE 20. PAIRED_FASTQ, Name of mate pair file if exists (Runs with failed mates will have a library layout of PAIRED but no paired fastq file) 21. WITHDRAWN, 0/1 to indicate if the file has been withdrawn, only present if a file has been withdrawn 22. WITHDRAWN_DATE, date of withdrawal, this should only be defined if a file is withdrawn 23. COMMENT, comment about reason for withdrawal 24. READ_COUNT, read count for the file 25. BASE_COUNT, basepair count for the file Any run_id can have up to 3 files associated with it. Single runs have one file. Paired runs can have anywhere from 1 to 3 files depending on the success of the pairing.