expinput kallisto folder empty

[Brian Estevez]

Hi,

I am using the latest version for PC  2.1.4.1

My goal is to do ICGS on single cell rna sequencing fastq files. From GUI, I first run raw sequence or processed on 6 fastq files: from 2 samples, each with index, cdna and cell barcode files). I do this in order to get  a Kalisto  expression input file and re-run for ICGS using process expression file mode. 

Oddly, there are two expression input folders being generated upon completion of first run on my fastq files:

1)  in the specified output  folder and another one  2)  alongside the fastq files.

There are 3 files in  '1)'  two folders that correspond to two of my samples containing files such as abundance.h5, abundance.tsv, and run_info.json, and a log.txt file.

There are no files in ' 2)' ExpressionInput/Kallisto

So I use the exp.txt file in Kalisto_Results folder. I run analysis and get the error: indexerror list inder out of range.

What should I try next?

-Brian

[Nathan Salomonis]

Hi Brian,

Can you send your log file from AltAnalyze and a screenshot of these directories? The default ICGS option for Minimum number of samples differing to 2 or 3 and correlation to 0.6 for this study design.

Best,

Nathan

--
You received this message because you are subscribed to the Google Groups "Alternative Splicing and Functional Prediction" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alt_predictio...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/alt_predictions/3ca4b90f-cbc3-44be-abe5-ac7fd5041581o%40googlegroups.com.

[Brian Estevez]

Attached are slides of individual screen shots, as well as individual screen shots. I think the main issue is my files are not being recognized and combined into single cell RNA sequencing data. 

It treats it as bulk sequencing and proceeds with bulk analysis. I can tell because the groups file shows only two columns but there should be several thousands.

Best,

-Brian

On Thursday, July 2, 2020 at 2:02:57 AM UTC-4, Nathan Salomonis wrote:

Hi Brian,

Can you send your log file from AltAnalyze and a screenshot of these directories? The default ICGS option for Minimum number of samples differing to 2 or 3 and correlation to 0.6 for this study design.

Best,

Nathan

On Thu, Jul 2, 2020 at 12:36 AM Brian Estevez <beste...@gmail.com> wrote:

Hi,

I am using the latest version for PC  2.1.4.1

My goal is to do ICGS on single cell rna sequencing fastq files. From GUI, I first run raw sequence or processed on 6 fastq files: from 2 samples, each with index, cdna and cell barcode files). I do this in order to get  a Kalisto  expression input file and re-run for ICGS using process expression file mode. 

Oddly, there are two expression input folders being generated upon completion of first run on my fastq files:

1)  in the specified output  folder and another one  2)  alongside the fastq files.

There are 3 files in  '1)'  two folders that correspond to two of my samples containing files such as abundance.h5, abundance.tsv, and run_info.json, and a log.txt file.

There are no files in ' 2)' ExpressionInput/Kallisto

So I use the exp.txt file in Kalisto_Results folder. I run analysis and get the error: indexerror list inder out of range.

What should I try next?

-Brian

--
You received this message because you are subscribed to the Google Groups "Alternative Splicing and Functional Prediction" group.

To unsubscribe from this group and stop receiving emails from it, send an email to alt_pre...@googlegroups.com.

[Brian Estevez]

Nathan,

Sorry. Also, here is the log file in addition to screenshot I sent earlier.. Let me know if its the correct log file.

On Thursday, July 2, 2020 at 2:02:57 AM UTC-4, Nathan Salomonis wrote:

Hi Brian,

Can you send your log file from AltAnalyze and a screenshot of these directories? The default ICGS option for Minimum number of samples differing to 2 or 3 and correlation to 0.6 for this study design.

Best,

Nathan

On Thu, Jul 2, 2020 at 12:36 AM Brian Estevez <beste...@gmail.com> wrote:

Hi,

I am using the latest version for PC  2.1.4.1

My goal is to do ICGS on single cell rna sequencing fastq files. From GUI, I first run raw sequence or processed on 6 fastq files: from 2 samples, each with index, cdna and cell barcode files). I do this in order to get  a Kalisto  expression input file and re-run for ICGS using process expression file mode. 

Oddly, there are two expression input folders being generated upon completion of first run on my fastq files:

1)  in the specified output  folder and another one  2)  alongside the fastq files.

There are 3 files in  '1)'  two folders that correspond to two of my samples containing files such as abundance.h5, abundance.tsv, and run_info.json, and a log.txt file.

There are no files in ' 2)' ExpressionInput/Kallisto

So I use the exp.txt file in Kalisto_Results folder. I run analysis and get the error: indexerror list inder out of range.

What should I try next?

-Brian

--
You received this message because you are subscribed to the Google Groups "Alternative Splicing and Functional Prediction" group.

To unsubscribe from this group and stop receiving emails from it, send an email to alt_pre...@googlegroups.com.

[Nathan Salomonis]

Hi Brian,

My apologies for the confusion. The current Kallisto version does not support Kallisto bus calls to produce sparse matrix results from 10x Genomics fastq files, only conventional bulk and fastq files where each encode for a single-cell (e.g., SMART-Seq) or sample. Hence, currently you would need to supply either:

1) A sparse matrix output from Cell Ranger (e.g., .mtx.gz or .h5)

2) dense tab-delimited matrix of counts 

3) dense tab-delimited matrix of scaled counts (log2 normalized, counts per 10,000 adjusted).

For an h5 file 

python AltAnalyze.py --platform RNASeq --species Hs --restrictBy protein_coding --excludeCellCycle no --removeOutliers yes --row_method hopach --ChromiumSparseMatrix /Users/GitHub/tests/demo_data/10X/input/cancer.h5 --output Users/GitHub/tests/demo_data/10X/input --runICGS yes  --expname cancer --downsample 2500

For a counts file

python AltAnalyze.py --platform RNASeq --species Hs --restrictBy protein_coding --excludeCellCycle no --removeOutliers yes --row_method hopach --ChromiumSparseMatrix /Users/GitHub/tests/demo_data/10X/input/cancer.h5 --output Users/GitHub/tests/demo_data/10X/input --runICGS yes  --expname cancer --dataFormat counts

You could also run Kallisto bus to get the input files, which I believe is fairly straight forward. Kallisto-bus is on our timeline but was delayed.

Best,

Nathan

To unsubscribe from this group and stop receiving emails from it, send an email to alt_predictio...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/alt_predictions/fded842a-fe43-45f1-ab61-b71a357b9bb9o%40googlegroups.com.