search

dereneaton / ipyrad

Interactive assembly and analysis of RAD-seq data sets
http://ipyrad.readthedocs.io
GNU General Public License v3.0
70 stars 39 forks source link

Step 3 gets to 85% and crashes. #437

Open Pgjhmb opened 3 years ago

Pgjhmb commented 3 years ago

Hello, I am running ipyrad on ezrad data for assemblies of consensus sequences within a single genus. It is believed that the species are distantly divergent. I have been running denovo without a reference on each of my samples, but each time (third time now) it has crashed after ~9 days, getting to the 85% point before doing so. Very frustrating.

Below is my params file. The cpu memory shouldn't be an issue.

I have 21 samples, between 2.5 million to 10 million reads of 300-750 bp long.

I would love any insight. I have used a whole month of crashed scripting.....

HALLLLP! Thanks. P

------- ipyrad params file (v.0.9.62)------------------------------------------- Tridacna-1_26_21 ## [0] [assembly_name]: Assembly name. Used to name output directories for assembly steps /data/paolo/Tridacna_mt_phylogeny/Paired_Reads ## [1] [project_dir]: Project dir (made in curdir if not present)

[2] [raw_fastq_path]: Location of raw non-demultiplexed fastq files

                           ## [3] [barcodes_path]: Location of barcodes file

/data/paolo/Tridacna_mt_phylogeny/Paired_Reads/*.fastq ## [4] [sorted_fastq_path]: Location of demultiplexed/sorted fastq files denovo ## [5] [assembly_method]: Assembly method (denovo, reference)

[6] [reference_sequence]: Location of reference sequence file

rad ## [7] [datatype]: Datatype (see docs): rad, gbs, ddrad, etc. GATC, ## [8] [restriction_overhang]: Restriction overhang (cut1,) or (cut1, cut2) 5 ## [9] [max_low_qual_bases]: Max low quality base calls (Q<20) in a read 33 ## [10] [phred_Qscore_offset]: phred Q score offset (33 is default and very standard) 6 ## [11] [mindepth_statistical]: Min depth for statistical base calling 6 ## [12] [mindepth_majrule]: Min depth for majority-rule base calling 10000 ## [13] [maxdepth]: Max cluster depth within samples 0.85 ## [14] [clust_threshold]: Clustering threshold for de novo assembly 0 ## [15] [max_barcode_mismatch]: Max number of allowable mismatches in barcodes 2 ## [16] [filter_adapters]: Filter for adapters/primers (1 or 2=stricter) 35 ## [17] [filter_min_trim_len]: Min length of reads after adapter trim 2 ## [18] [max_alleles_consens]: Max alleles per site in consensus sequences 0.05 ## [19] [max_Ns_consens]: Max N's (uncalled bases) in consensus 0.05 ## [20] [max_Hs_consens]: Max Hs (heterozygotes) in consensus 4 ## [21] [min_samples_locus]: Min # samples per locus for output 0.2 ## [22] [max_SNPs_locus]: Max # SNPs per locus 8 ## [23] [max_Indels_locus]: Max # of indels per locus 0.5 ## [24] [max_shared_Hs_locus]: Max # heterozygous sites per locus 0, 0, 0, 0 ## [25] [trim_reads]: Trim raw read edges (R1>, <R1, R2>, <R2) (see docs) 0, 0, 0, 0 ## [26] [trim_loci]: Trim locus edges (see docs) (R1>, <R1, R2>, <R2) p, s, l ## [27] [output_formats]: Output formats (see docs)

[28] [pop_assign_file]: Path to population assignment file

                           ## [29] [reference_as_filter]: Reads mapped to this reference are removed in step 3
isaacovercast commented 3 years ago

Hello P, What makes you think the memory isn't an issue? The number of reads per sample and the locus length will both contribute to consume a TON of RAM during step 3. How much ram do you have and how many cores? Also, what is the exact text of the error message?

Pgjhmb commented 3 years ago

Hello,

Thanks for getting back ot me, We have 256gb of RAM and htop does not indicate we are going into swap. Not even using half the ram. We have 32 cores, and I allocated 19 of them. The server has crashed all 3 times and I am starting to think it is not a coincidence.

Thanks for any insight you can provide. I am trying to get some assemblies. Any other programs you recommend I would be interested in hearing. I might set up individual runs with the specific species reference. Paolo

On Sun, Feb 14, 2021 at 10:16 AM Isaac Overcast notifications@github.com wrote:

Hello P, What makes you think the memory isn't an issue? The number of reads per sample and the locus length will both contribute to consume a TON of RAM during step 3. How much ram do you have and how many cores? Also, what is the exact text of the error message?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dereneaton/ipyrad/issues/437#issuecomment-778835443, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOXHCLVCOKCC4O5MTC4BQNDS7AVSTANCNFSM4XTRB3KA .

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

isaacovercast commented 3 years ago

Hey Paolo, Can you show me the error message from when it crashes? -isaac

isaacovercast commented 3 years ago

The long reads can also cause massive problems during clustering and alignment given that the distal ends of R1 and R2 can obtain very high error rates. You might consider looking at the results of fastqc for a couple of your samples and using the trim_reads parameter in step 2 to trim off the regions with very low base quality.

Pgjhmb commented 3 years ago

Hey Issac,

The internet connection has gone down at the lab (HIMB), so I will send it when once I have access.

Thanks

On Sun, Feb 14, 2021 at 10:26 AM Isaac Overcast notifications@github.com wrote:

The long reads can also cause massive problems during clustering and alignment given that the distal ends of R1 and R2 can obtain very high error rates. You might consider looking at the results of fastqc for a couple of your samples and using the trim_reads parameter in step 2 to trim off the regions with very low base quality.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dereneaton/ipyrad/issues/437#issuecomment-778836701, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOXHCLQOC3XSDP5MUK6QLFLS7AWW3ANCNFSM4XTRB3KA .

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

Pgjhmb commented 3 years ago

Hey Isaac,

I can't confirm that it failed due to the script or the power supply. We have UPSʻs for this very reason, and had them recently replaced. What are the file outputs after step 3? I have the following folders: Tridacna-tmpalign, Tridacna_clust_0.85, and Tridacna_edits.

There is no log file, nor an error message. Because the server crashes while using a GNU screen, so it never displays the error log. Is there a way to write within the params file the creation of a log file and an error file?

Thanks, Paolo

On Sun, Feb 14, 2021 at 10:37 AM Paolo Marra-Biggs paolomb@hawaii.edu wrote:

Hey Issac,

The internet connection has gone down at the lab (HIMB), so I will send it when once I have access.

Thanks

On Sun, Feb 14, 2021 at 10:26 AM Isaac Overcast notifications@github.com wrote:

The long reads can also cause massive problems during clustering and alignment given that the distal ends of R1 and R2 can obtain very high error rates. You might consider looking at the results of fastqc for a couple of your samples and using the trim_reads parameter in step 2 to trim off the regions with very low base quality.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dereneaton/ipyrad/issues/437#issuecomment-778836701, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOXHCLQOC3XSDP5MUK6QLFLS7AWW3ANCNFSM4XTRB3KA .

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

isaacovercast commented 3 years ago

Wait, when you say "it has crashed" you mean the server crashes? I thought you were talking about the ipyrad process. If the server is crashing that is a hardware issue my friend.

Did you consider this: "You might consider looking at the results of fastqc for a couple of your samples and using the trim_reads parameter in step 2 to trim off the regions with very low base quality."

Pgjhmb commented 3 years ago

Hey Isaac, I tried taking a subset of the samples and just use 1 library, using the same parameters as listed before and I get this error.

Encountered an Error.

Message: IPyradError: [bwa_index] Pack FASTA... [gzread] Is a directory

Then when I tried using a denovo+reference assembly method, I get this error.

Encountered an Error.

Message: datatype + assembly_method combo not currently supported.

I have also tried to trim the reads beforehand using BBduk and trimming based on a quality score.

Any help would be appreciated, Paolo

On Sun, Feb 14, 2021 at 9:10 PM Paolo Marra-Biggs paolomb@hawaii.edu wrote:

Hey Isaac,

I can't confirm that it failed due to the script or the power supply. We have UPSʻs for this very reason, and had them recently replaced. What are the file outputs after step 3? I have the following folders: Tridacna-tmpalign, Tridacna_clust_0.85, and Tridacna_edits.

There is no log file, nor an error message. Because the server crashes while using a GNU screen, so it never displays the error log. Is there a way to write within the params file the creation of a log file and an error file?

Thanks, Paolo

On Sun, Feb 14, 2021 at 10:37 AM Paolo Marra-Biggs paolomb@hawaii.edu wrote:

Hey Issac,

The internet connection has gone down at the lab (HIMB), so I will send it when once I have access.

Thanks

On Sun, Feb 14, 2021 at 10:26 AM Isaac Overcast notifications@github.com wrote:

The long reads can also cause massive problems during clustering and alignment given that the distal ends of R1 and R2 can obtain very high error rates. You might consider looking at the results of fastqc for a couple of your samples and using the trim_reads parameter in step 2 to trim off the regions with very low base quality.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/dereneaton/ipyrad/issues/437#issuecomment-778836701, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOXHCLQOC3XSDP5MUK6QLFLS7AWW3ANCNFSM4XTRB3KA .

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

-- Paolo Marra-Biggs University of Hawaii at Mānoa Marine Biology MSc. Student ToBo Lab Phone: 1 (831) 254-3780

isaacovercast commented 3 years ago

Regarding my previous message: Wait, when you say "it has crashed" you mean the server crashes? I thought you were talking about the ipyrad process. If the server is crashing that is a hardware issue my friend. <- Was the original crash a hardware failure?

If you want help you need to provide more specific information, like "what samples did you retain for the single library analysis" "what parameters did you use in your params file" what was the step the assembly crashed in, what was all of the output from the assembly run and all of the output of the error message.

When you say you tried to trim the reads beforehand, how did you do this, what were the exact settings you used for trimming, did you run ipyrad after trimming and if so what happened and how was it different from pre-trimming behavior.

This message is self-explanatory: "Message: datatype + assembly_method combo not currently supported."

This is evolving away from a very specific error in ipyrad to more of a help thread, which is more appropriate for the gitter channel: https://gitter.im/dereneaton/ipyrad