cellranger count error

Question

[connorjfausto]

Hello!

I am new to processing snRNAseq data, and am going through the 10X cellranger pipeline.

I could not find any bam files to download from the SRR numbers, so I chose to process raw fastq files.

This is how I downloaded the files, using 1 as an example:

1. prefetch SRA files

prefetch SRR9141212

1. Fastq-dump zipped files

   fastq-dump --split-3 --skip-technical --readids --read-filter pass --gzip /project/nilslind_369/cfausto/wilson-diabetes/input/pretrim/SRR9141212.sra

2. Rename the two fastq.gz files to match cellranger count nomenclature (moved into one directory)>

   mv SRR9141212_pass_1.fastq.gz ctrl1_S1_L001_R1_001.fastq.gz

SRR9141212_pass_2.fastq.gz ctrl1_S1_L001_R2_001.fastq.gz

1. Script for cellranger count

   cellranger count --id=ctrl1 \ --transcriptome=/project/nilslind_369/cfausto/Ref_Genome/hg38/Homo-sapien_genome \ --fastqs=/project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1

Here is the error message when the job fails early on:

FASTQ header mismatch detected at line 4 of input files "/project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1/ctrl1_S1_L001_R1_001.fastq.gz" and "/project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1/ctrl1_S1_L001_R2_001.fastq.gz": file: "/project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1/ctrl1_S1_L001_R1_001.fastq.gz", line: 4

When I head the fastq file, it looks like it's in the wrong format:

oߎ#??$x?~?4?xr/?/9??~?????O?8??o?<?u?y?s?<ߐy???ݩ??'???,?o?{??Szeu??????????_??????????VJ??????׭??~Q??z????????Z?T??W??Z??????????????~???w????V??????????uY㕿????g?M?g??wu???/???Ϫ????r?W?ˌ????*???????G5?S2 ???k?N7&?????me?*(?ɿVg<Qȳk~\u?????????????}?xM????~:)ڿߝf?_??????{~?.1?H??wr??????)m?㱮?@?s~~?+d}???<Mt??????j<_??_?_k?3]??S|3?y??[?H??M@^u}?s??[?_??????F??Q??????o?^???ʣ?T9????R?{~h??'s^?u6??Ϳ?o??o?2o?{<r?'rZ???J???_?????a??Gd?y?tH+V?y?? ?ߺ Ug???c??K\?u???n??ư>G??!?oÃ?ux?????4??H?b??j?S[ϯ?+??<gu5י>o???\[?=????1?;?naVg??F??uZ?a? ??F?c?je?{>|? ??wu??_?f????"?1????W????<??????S?#??D? ?.??0??)x]??VS?????m?g|?<???98?c?.?gC'?<c c[?S????Qob????????s??Y??????L??c???????p[?? =(u?u?V?Z??0?????U?/?W?t΃?-?v0?k??1?????v?yf??? S????Z????Z_.؝u<????T'?ǜ,???it7?c?????0?????P??җ*?y?+:?xoQZ??*?????n?9Yއs sPu+?ʕ??>?x?Ї??,?????g]???)?Ū?GU?6=????W?I?^?gG??????+8C <.??4?RKu?t???繩? |?M??5?O=?? :?? ?? ???q??D? ?n???HѵY}?q???.????s???_+W=S?HH??F???5$m?go???0E??Z?H= ??y֭??????챭??:?M?n?s??Ζ?1O????y?/??kZ?z?x???)-?????ş????pO?pBգB [3}p͍ї??H???1'?Ӿ?|? ?a1?? ]?ސ~?l ??? Lϐ??q?x???0???!>?b?y??.??3}I)?A?RlhhO?|???p??nc?> /? |j<??*?:-?4DЊΎ?ng????? ?-?aٲ?<?4??????go????튇??TKc䝓?X?5D??V??+? B?v??gl??????<{C?o?<?U?? Z??Bh??W??j?Sp8??"C%?????ˈ??3M???go???????8?:?+ˆ??x???X?'?????3?7?????z????H>?p?``E̯?] }?x??????_"???>??d?b*賍X??l?????J??,?Bi??]????go??????Ly?????????y??>>ݺ??3h?v?CQ?f?0????S"?R?y???d?????3??5?w??c??>??O¤?n=?1hO/Rp? D'xe?F????/????Z?I??@?]???1??Ї?Hj ^ ??8???Z?bɊ?oF?LZ????ME8?g?oanj/?y??>?F?OavO%?F??!"ţ!??D?.??!M??1'?o?7? ]??!??xg]?:+C?:?-?Y??ND?u?/??<?d???,^???k??t? ?72{C???y.86?-WR??9?^"?b?? Z@??3G??7a]?? j???O?n s?9???Sy\?>?vY??cN?G ??h??????z?????En?q]??o?G????dy{????? ???I???<??;N1wsUB??b?9??]??Ni?G%$j???%??-?k?=???|?<?dy??/?y?????{??W??j=- ?W?r?~?<S??˶?$?r???cN?|?]??n̳7t]{CL?kΝ0 ???~l?<l??Q?K??GQd?$>???Kh?}???S|?:{CW? ??(?1̀???o1?k9?y̳7t??0aNH@x?2?? 6t?????H:P??ƙO?z??K?? ? e˯)no=???n??Z=?dy??^;?z?|??????i*脓???W????????V?Sx? ?????<{C??ÌO??撬?l?8N??ې?m~? ?8<??[?zvs?<????G?=?<??]??R?2d?iV???bT?8???1??(+n&NM???O?????<{C???n??P^.? ?𭱀[4S?f??.??[:?Hb%???? ????<{C??7D?&?Q?f???B?os]??#.??:O`???4???ݕI<!?/?P?????n^????s

I'm not sure why it looks like this, maybe because it's a zipped file? Any different way to get the fastq files would be greatly appreciated!

[GenoMax]

FASTQ header mismatch detected at line 4 of input files

If the files are out of sync you can use repair.sh from BBMap suite to bring the files back in sync removing any singleton reads.

repair.sh in1=broken1.fq.gz in2=broken2.fq.gz out1=fixed1.fq.gz out2=fixed2.fq.gz outs=singletons.fq.gz repair 

[Pierre Lindenbaum]

what is the output of

gunzip -c /project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1/ctrl1_S1_L001_R1_001.fastq.gz | head

and

gunzip -c /project/nilslind_369/cfausto/wilson-diabetes/input/ctrl1/ctrl1_S1_L001_R2_001.fastq.gz | head

[connorjfausto]

The outputs look like normal fastq files:

R1

@SRR9141212.1.1 1 length=26
NCTGCTAAGCTTTGTTTCCTAATTAA
+SRR9141212.1.1 1 length=26
#FFFFFFFFFFFFFFFFF:FFFFFFF
@SRR9141212.2.1 2 length=26
NGGTCATCTTTCCTCTCCTTAATACT
+SRR9141212.2.1 2 length=26
#FFF:FFFFFFFFFFFF:FFFFFFFF
@SRR9141212.3.1 3 length=26
NGTATTGCATGCCACGCTTTTGTCCG

R2

@SRR9141212.1.2 1 length=91
GGAAGCAATTTTGGAGGTGGTGGAAGCTACAATGATTTTGGGAATTACAACAATCAGTCTTCAAATTTTGGACCCATGAAGGGAGGAAATT
+SRR9141212.1.2 1 length=91
FFF:FFF,FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFF:FFF:FFF
@SRR9141212.2.2 2 length=91
GTGCTCTTTTAGCTGTTCTTAGGTAGCTCGTCTGGTTTCGGGGGTCTTAGCTTTGGCTCTCCTTGCAAAGTTATTTCTAGTTAATTCATTA
+SRR9141212.2.2 2 length=91
FFFFF::FFFFFFFF,FFFFFF:FFFFFFFFFFFFF:FF:FF,FFFFFFFFFFFFFFFFFF::FFFF,FF:F:FFFFFFF::FFFFFFF:F
@SRR9141212.3.2 3 length=91
GGTCCTGCAGGAATAAGCCCTCAGCACCTACGGAGCTGGTAGAATCTGTCCATATTGCACGTTCCCCTTGCAACTCCATACCATCTAACAC

[GenoMax]

Download fastq files from this link: https://trace.ncbi.nlm.nih.gov/Traces/index.html?view=run_browser&acc=SRR9141212&display=download

[connorjfausto]

I'm wondering how this would be different than downloading with fastq-dump because it's a 58GB file, which I wouldn't necessarily want on my personal computer

[GenoMax]

Dumping 10x data via sratools has caused issues in past. If you are sure the file you have is ok then carry on.