raphassaraf/MNLP_M3_document_encoder_QA
048
1---2tags:3- sentence-transformers4- sentence-similarity5- feature-extraction6- generated_from_trainer7- dataset_size:968- loss:MultipleNegativesRankingLoss9base_model: BAAI/bge-small-en-v1.510widget:11- source_sentence: What are the de facto required fields in a SAM/BAM read group?12 sentences:13 - "Question: Several gene set enrichment methods are available, the most famous/popular\14 \ is the Broad Institute tool. Many other tools are available (See for example\15 \ the biocView of GSE which list 82 different packages). There are several parameters\16 \ in consideration :\n\nthe statistic used to order the genes, \nif it competitive\17 \ or self-contained,\nif it is supervised or not,\nand how is the enrichment score\18 \ calculated.\n\nI am using the fgsea - Fast Gene Set Enrichment Analysis package\19 \ to calculate the enrichment scores and someone told me that the numbers are\20 \ different from the ones on the Broad Institute despite all the other parameters\21 \ being equivalent.\nAre these two methods (fgsea and Broad Institute GSEA) equivalent\22 \ to calculate the enrichment score?\nI looked to the algorithms of both papers,\23 \ and they seem fairly similar, but I don't know if in real datasets they are\24 \ equivalent or not.\nIs there any article reviewing and comparing how does the\25 \ enrichment score method affect to the result?\n\nAnswer: According to the FGSEA\26 \ preprint:\n\nWe ran reference GSEA with default parameters. The permutation\27 \ number\n was set to 1000, which means that for each input gene set 1000\n \28 \ independent samples were generated. The run took 100 seconds and\n resulted\29 \ in 79 gene sets with GSEA-adjusted FDR q-value of less than\n 10−2. All significant\30 \ gene sets were in a positive mode. First, to get\n a similar nominal p-values\31 \ accuracy we ran FGSEA algorithm on 1000\n permutations. This took 2 seconds,\32 \ but resulted in no significant hits\n due after multiple testing correction\33 \ (with FRD ≤ 1%).\n\nThus, FGSEA and GSEA are not identical.\nAnd again in the\34 \ conclusion:\n\nConsequently, gene sets can be ranked more precisely in the results\n\35 \ and, which is even more important, standard multiple testing\n correction\36 \ methods can be applied instead of approximate ones as in\n [GSEA].\n\nThe author\37 \ argues that FGSEA is more accurate, so it can't be equivalent.\nIf you are interested\38 \ specifically in the enrichment score, that was addressed by the author in the\39 \ preprint comments:\n\nValues of enrichment scores and normalized enrichment\40 \ scores are the\n same for both broad version and fgsea.\n\nSo that part seems\41 \ to be the same."42 - 'Question: I am running samtools mpileup (v1.4) on a bam file with very choppy43 coverage (ChIP-seq style data). I want to get a first-pass list of positions with44 SNVs and their frequency as reported by the read counts, but no matter what I45 do, I keep getting all SNVs filtered out as not passing QC.46 47 What''s the magic parameter set for an initial list of SNVs and frequencies?48 49 EDIT: this is a question I posted on "the other" website, but didn''t get a reply50 there.51 52 53 Answer: I used this in the past for ChIP-seq data and it generated SNVs:54 55 samtools mpileup \56 57 --uncompressed --max-depth 10000 --min-MQ 20 --ignore-RG --skip-indels \58 59 --fasta-ref ref.fa file.bam \60 61 | bcftools call --consensus-caller \62 63 > out.vcf64 65 66 This was samtools 1.3 in case that makes a difference.'67 - "Question: The SAM specification indicates that each read group must have a unique\68 \ ID field, but does not mark any other field as required. \nI have also discovered\69 \ that htsjdk throws exceptions if the sample (SM) field is empty, though there\70 \ is no indication in the specification that this is required. \nAre there other\71 \ read group fields that I should expect to be required by common tools? \n\n\72 Answer: The sample tag (i.e. SM) was a mandatory tag in the initial SAM spec (see\73 \ the .pages file; you need a mac to open it). When transitioned to Latex, this\74 \ requirement was mysteriously dropped. Picard is conforming to the initial spec.\75 \ Anyway, the sample tag is important to quite a few tools. I would encourage\76 \ you to add it."77- source_sentence: Is the optional SAM NM field strictly computable from the MD and78 CIGAR?79 sentences:80 - "Question: I'm looking for tools to check the quality of a VCF I have of a human\81 \ genome. I would like to check the VCF against publicly known variants across\82 \ other human genomes, e.g. how many SNPs are already in public databases, whether\83 \ insertions/deletions are at known positions, insertion/deletion length distribution,\84 \ other SNVs/SVs, etc.? I suspect that there are resources from previous projects\85 \ to check for known SNPs and InDels by human subpopulations.\nWhat resources\86 \ exist for this, and how do I do it? \n\nAnswer: To achieve (at least some of)\87 \ your goals, I would recommend the Variant Effect Predictor (VEP). It is a flexible\88 \ tool that provides several types of annotations on an input .vcf file. I agree\89 \ that ExAC is the de facto gold standard catalog for human genetic variation\90 \ in coding regions. To see the frequency distribution of variants by global\91 \ subpopulation make sure \"ExAC allele frequencies\" is checked in addition to\92 \ the 1000 genomes. \nOutput in the web-browser:\n\nIf you download the annotated\93 \ .vcf, frequencies will be in the INFO field:\n##INFO=<ID=CSQ,Number=.,Type=String,Description=\"\94 Consequence annotations from Ensembl VEP. Format: Allele|Consequence|IMPACT|SYMBOL|Gene|Feature_type|Feature|BIOTYPE|EXON|INTRON|HGVSc|HGVSp|cDNA_position|CDS_position|Protein_position|Amino_acids|Codons|Existing_variation|DISTANCE|STRAND|FLAGS|SYMBOL_SOURCE|HGNC_ID|TSL|SIFT|PolyPhen|AF|AFR_AF|AMR_AF|EAS_AF|EUR_AF|SAS_AF|AA_AF|EA_AF|ExAC_AF|ExAC_Adj_AF|ExAC_AFR_AF|ExAC_AMR_AF|ExAC_EAS_AF|ExAC_FIN_AF|ExAC_NFE_AF|ExAC_OTH_AF|ExAC_SAS_AF|CLIN_SIG|SOMATIC|PHENO|MOTIF_NAME|MOTIF_POS|HIGH_INF_POS|MOTIF_SCORE_CHANGE\n\95 \nThe previously mentioned Annovar can also annotate with ExAC allele frequencies.\96 \ Finally, should mention the newest whole-genome resource, gnomAD."97 - 'Question: I produced a bam file by aligning reads to a small set of synthetic98 sequences using bwa-mem.99 100 I am heavily filtering reads that are not paired and of a certain orientation.101 102 Applying the filtering, I get a few thousands of reads:103 104 samtools view -h $myfilebam | \105 106 samtools view -h -F4 - | \107 108 samtools view -h -F8 - | \109 110 samtools view -h -F256 - | \111 112 samtools view -h -F512 - | \113 114 samtools view -h -F1024 - | \115 116 samtools view -h -F2048 - | \117 118 samtools view -h -f16 - | \119 120 samtools view -h -f32 - | wc -l121 122 123 Gives me 89502 reads.124 125 If I then pipe this into samtools mpileup, I get no results:126 127 samtools view -h $myfilebam | \128 129 samtools view -h -F4 - | \130 131 samtools view -h -F8 - | \132 133 samtools view -h -F256 - | \134 135 samtools view -h -F512 - | \136 137 samtools view -h -F1024 - | \138 139 samtools view -h -F2048 - | \140 141 samtools view -h -f16 - | \142 143 samtools view -h -f32 - | \144 145 samtools mpileup --excl-flags 0 -Q0 -B -d 999999 - | wc -l146 147 148 Returns 0.149 150 I tried different combinations of filtering, and when I do both -f 16 and -f 32151 returns empty, but if I do either of those, then it works:152 153 samtools view -h $myfilebam | \154 155 samtools view -h -F4 - | \156 157 samtools view -h -F8 - | \158 159 samtools view -h -F256 - | \160 161 samtools view -h -F512 - | \162 163 samtools view -h -F1024 - | \164 165 samtools view -h -F2048 - | \166 167 samtools view -h -f16 - | \168 169 samtools mpileup --excl-flags 0 -Q0 -B -d 999999 - | wc -l170 171 172 Returns 1056.173 174 Any ideas why? My thinking was that it would work with --excl-flags 0.175 176 EDIT: substituting mpileup for depth does work, and prints out each position and177 the depth as expected.178 179 EDIT2: adding -q 0 to mpileup gives the same empty result.180 181 Thanks in advance182 183 184 Answer: By using -h in the samtools view command, you''re including all the header185 lines in your word count. If you happen to have about 89500 reference sequences,186 then the lengths of those would all appear in the header and inflate the -h word187 count, but not the mpileup count. Try piping it through an additional samtools188 view (i.e. without -h) and see if the counts change:189 190 ...191 192 samtools view -h -f32 - | \193 194 samtools view | wc -l195 196 197 Also, samtools mpileup by default only considers high-quality bases and concordant198 reads. Try adding a -A to your mpileup line (which stops anomalous read pairs199 from being discarded):200 201 ...202 203 samtools mpileup -A -Q0 -B -d 999999 - | wc -l204 205 206 Whether or not this is actually a good idea will be dependent on what you want207 to get out of the analysis, and what the downstream programs / analyses are expecting.'208 - "Question: From SAM Optional Fields Specification the NM field is \n\nEdit distance\209 \ to the reference, including ambiguous bases but excluding clipping\n\nAssuming\210 \ both the MD and CIGAR are present, is the edit distance simply the number of\211 \ characters [A-Z] appearing in the MD field plus the number of bases inserted\212 \ (xI, if any) from the CIGAR string? Are there any other complications? \n\n\213 Answer: Assuming both the MD and CIGAR are present and correct, then yes, you\214 \ can parse both to get the edit distance (NM auxiliary tag). One big caveat to\215 \ this is that there's a reason that the samtools calmd command exists, since\216 \ it's historically been the case that not all aligners have output correct MD\217 \ strings. It's rare for the CIGAR string to be wrong and that'd be more of a\218 \ catastrophic error on the part of an aligner. For what it's worth, if the NM\219 \ auxiliary is absent on a given alignment but present on others produced by the\220 \ same aligner then it's fair to assume NM:i:0 for a given alignment by default\221 \ (many aligners only produce NM:i:XXX if the edit distance is at least 1)."222- source_sentence: How to read structural variant VCF?223 sentences:224 - "Question: I am calling SNPs from WGS samples produced at my lab. I am currently\225 \ using bwa-mem for mapping Illumina reads as it is recommended by GATK best practice.\226 \ However, bwa is a bit slow. I heard from my colleague that SNAP is much faster\227 \ than bwa. I tried it on a small set of reads and it is indeed faster. However,\228 \ I am not sure how it works with downstream SNP callers, so here are my questions:\229 \ have you used SNAP for short-read mapping? What is your experience? Does SNAP\230 \ work well with SNP callers like GATK and freebayes? Thanks!\n\nAnswer: GATK\231 \ best practices are explicably meant to consume BWA MEM generated BAM. Whilst\232 \ SNAP may be faster, the Broad will not have tested it for compatibility with\233 \ GATK as such you can't guaranty using it won't have unexpected consequences.\234 \ \nAs such you'd be better off using BWA MEM because I assume accurately called\235 \ variation is always better than fast and incorrectly called variation. The\236 \ main issue you'll have is ensuring shorter split hits and mapping quality are\237 \ reported in the same way as bwa MEM -M which GATK/Picard is expecting. Ultimately\238 \ however you'd be better off posting this question on the GATK forum. \nIt's\239 \ also worth noting that the soon to be released GATK 4 will utilise bwaspark\240 \ which can distribute it's alignment processes across Apache Spark for increase\241 \ performance. Consequently I can't see SNAP being adopted anytime soon."242 - 'Question: I have a computer engineering background, not biology.243 244 I started working on a bioinformatics project recently, which involves de-novo245 assembly. I came to know the terms Transcriptome and Genome, but I cannot identify246 the difference between these two.247 248 I know a transcriptome is the set of all messenger RNA molecules in a cell, but249 am not sure how this is different from a genome.250 251 252 Answer: In brief, the “genome” is the collection of all DNA present in the nucleus and the mitochondria253 of a somatic cell. The initial product of genome expression is the “transcriptome”,254 a collection of RNA molecules derived from those genes.'255 - "Question: The IGSR has a sample for encoding structural variants in the VCF 4.0\256 \ format.\nAn example from the site (the first record):\n#CHROM POS ID REF\257 \ ALT QUAL FILTER INFO FORMAT NA00001\n1 2827693 . CCGTGGATGCGGGGACCCGCATCCCCTCTCCCTTCACAGCTGAGTGACCCACATCCCCTCTCCCCTCGCA\258 \ C . PASS SVTYPE=DEL;END=2827680;BKPTID=Pindel_LCS_D1099159;HOMLEN=1;HOMSEQ=C;SVLEN=-66\259 \ GT:GQ 1/1:13.9\n\nHow to read it? From what I can see:\n\nThis is a deletion\260 \ (SVTYPE=DEL)\nThe end position of the variant comes before the starting position\261 \ (reverse strand?)\nThe reference starts from 2827693 to 2827680 (13 bases on\262 \ the reverse strand)\nThe difference between reference and alternative is 66\263 \ bases (SVLEN=-66)\n\nThis doesn't sound right to me. For instance, I don't see\264 \ where exactly the deletion starts. The SVLEN field says 66 bases deleted, but\265 \ where? 2827693 to 2827680 only has 13 bases between.\nQ: How to read the deletion\266 \ correctly from this structural VCF record? Where is the missing 66-13=53 bases?\n\267 \nAnswer: I just received a reply from 1000Genomes regarding this. I'll post it\268 \ in its entirety below:\n\nLooking at the example you mention, I find it difficult\269 \ to come up with an\n interpretation of the information whereby the stated end\270 \ seems to be correct,\n so believe that this may indeed be an error.\nSince\271 \ the v4.0 was created, however, new versions of VCF have been introduced,\n \272 \ improving and correcting the specification. The current version is v4.3\n (http://samtools.github.io/hts-specs/).\273 \ I believe the first record shown on\n page 11 provides an accurate example\274 \ of this type of deletion.\nI will update the web page to include this information.\n\275 \nSo we can take this as official confirmation that we were all correct in suspecting\276 \ the example was just wrong."277- source_sentence: Publicly available genome sequence database for viruses?278 sentences:279 - "Question: This question is based on a question on BioStars posted >2 years ago\280 \ by user jack.\nIt describes a very frequent problem of generating GO annotations\281 \ for non-model organisms. While it is based on some specific format and single\282 \ application (Ontologizer), it would be useful to have a general description\283 \ of the pathway to getting to a GAF file. \nNote, that the input format is lacking\284 \ a bit of essential information, like how it was obtained. Therefore, it is har\285 \ to assign evidence code. Therefore, lets assume that the assignments of GO terms\286 \ were done automagically. \n\nI want to do the Gene enrichment using Ontologizer\287 \ without a\n predefined association file(it's not model organism). \nI have\288 \ parsed a file with two columns for that organism like this : \ngeneA GO:0006950,GO:0005737\n\289 geneB GO:0016020,GO:0005524,GO:0006468,GO:0005737,GO:0004674,GO:0006914,GO:0016021,GO:0015031\n\290 geneC GO:0003779,GO:0006941,GO:0005524,GO:0003774,GO:0005516,GO:0005737,GO:0005863\n\291 geneD GO:0005634,GO:0003677,GO:0030154,GO:0006350,GO:0006355,GO:0007275,GO:0030528\n\292 \nI have downloaded the .ob file from Gene ontology file which contain\n this\293 \ information (from here) : \n!\n! GO IDs (primary only) and name text strings\n\294 ! GO:0000000 [tab] text string [tab] F|P|C\n! where F = molecular function, P\295 \ = biological process, C = cellular component\n!\nGO:0000001 mitochondrion inheritance\296 \ P\nGO:0000002 mitochondrial genome maintenance P\nGO:0000003 reproduction\297 \ P\nGO:0000005 ribosomal chaperone activity F\nGO:0000006 high affinity\298 \ zinc uptake transmembrane transporter activity F\nGO:0000007 low-affinity\299 \ zinc ion transmembrane transporter activity F\nGO:0000008 thioredoxin F\n\300 GO:0000009 alpha-1,6-mannosyltransferase activity F\nGO:0000010 trans-hexaprenyltranstransferase\301 \ activity F\nGO:0000011 vacuole inheritance P\n\nWhat I need as output is\302 \ .gaf file in the following format (in the\n format of the files here):\n!gaf-version:\303 \ 2.0\n\n!Project_name: Leishmania major GeneDB\n\n!URL: http://www.genedb.org/leish\n\304 \n!Contact Email: mb4@sanger.ac.uk\n\n GeneDB_Lmajor LmjF.36.4770 LmjF.36.4770\305 \ GO:0003723 PMID:22396527 ISO GeneDB:Tb927.10.10130 F mitochondrial\306 \ RNA binding complex 1 subunit, putative LmjF36.4770 gene taxon:347515\307 \ 20120910 GeneDB_Lmajor \n GeneDB_Lmajor LmjF.36.4770 LmjF.36.4770\308 \ GO:0044429 PMID:20660476 ISS C mitochondrial RNA binding\309 \ complex 1 subunit, putative LmjF36.4770 gene taxon:347515 20100803\310 \ GeneDB_Lmajor GeneDB_Lmajor LmjF.36.4770 LmjF.36.4770 \311 \ GO:0016554 PMID:22396527 ISO GeneDB:Tb927.10.10130 P mitochondrial\312 \ RNA binding complex 1 subunit, putative LmjF36.4770 gene taxon:347515\313 \ 20120910 GeneDB_Lmajor \n GeneDB_Lmajor LmjF.36.4770 LmjF.36.4770\314 \ GO:0048255 PMID:22396527 ISO GeneDB:Tb927.10.10130 P mitochondrial\315 \ RNA binding complex 1 subunit, putative LmjF36.4770 gene taxon:347515\316 \ 20120910 GeneDB_Lmajor \n\nHow to create your own GO association file\317 \ (gaf)?\n\nAnswer: Here's a Perl script that can do this:\n#!/usr/bin/env perl\318 \ \nuse strict;\nuse warnings;\n\n## Change this to whatever taxon you are working\319 \ with\nmy $taxon = 'taxon:1000';\nchomp(my $date = `date +%Y%M%d`);\n\nmy (%aspect,\320 \ %gos);\n## Read the GO.terms_and_ids file to get the aspect (sub ontology)\n\321 ## of each GO term. \nopen(my $fh, $ARGV[0]) or die \"Need a GO.terms_and_ids\322 \ file as 1st arg: $!\\n\";\nwhile (<$fh>) {\n next if /^!/;\n chomp;\n\323 \ my @fields = split(/\\t/);\n ## $aspect{GO:0000001} = 'P'\n $aspect{$fields[0]}\324 \ = $fields[2];\n}\nclose($fh);\n\n## Read the list of gene annotations\nopen($fh,\325 \ $ARGV[1]) or die \"Need a list of gene annotattions as 2nd arg: $!\\n\";\nwhile\326 \ (<$fh>) {\n chomp;\n my ($gene, @terms) = split(/[\\s,]+/);\n ## $gos{geneA}\327 \ = (go1, go2 ... goN)\n $gos{$gene} = [ @terms ];\n}\nclose($fh);\n\nforeach\328 \ my $gene (keys(%gos)) {\n foreach my $term (@{$gos{$gene}}) {\n ##\329 \ Warn and skip if there is no aspect for this term\n if (!$aspect{$term})\330 \ {\n print STDERR \"Unknown GO term ($term) for gene $gene\\n\";\n\331 \ next;\n }\n ## Build a pseudo GAF line \n my\332 \ @out = ('DB', $gene, $gene, ' ', $term, 'PMID:foo', 'TAS', ' ', $aspect{$term},\n\333 \ $gene, ' ', 'protein', $taxon, $date, 'DB', ' ',\334 \ ' ');\n print join(\"\\t\", @out). \"\\n\";\n }\n}\n\nMake it executable\335 \ and run it with the GO.terms_and_ids file as the 1st argument and the list of\336 \ gene annotations as the second. Using the current GO.terms_and_ids and the example\337 \ annotations in the question, I get:\n$ foo.pl GO.terms_and_ids file.gos \nDB\338 \ geneD geneD GO:0005634 PMID:foo TAS C geneD protein\339 \ taxon:1000 20170308 DB \nDB geneD geneD GO:0003677 PMID:foo\340 \ TAS F geneD protein taxon:1000 20170308 DB \nDB geneD\341 \ geneD GO:0030154 PMID:foo TAS P geneD protein taxon:1000\342 \ 20170308 DB \nUnknown GO term (GO:0006350) for gene geneD\nDB geneD\343 \ geneD GO:0006355 PMID:foo TAS P geneD protein taxon:1000\344 \ 20170308 DB \nDB geneD geneD GO:0007275 PMID:foo TAS\345 \ P geneD protein taxon:1000 20170308 DB \nDB geneD geneD\346 \ GO:0030528 PMID:foo TAS F geneD protein taxon:1000 20170308\347 \ DB \nDB geneB geneB GO:0016020 PMID:foo TAS C geneB\348 \ protein taxon:1000 20170308 DB \nDB geneB geneB GO:0005524\349 \ PMID:foo TAS F geneB protein taxon:1000 20170308 DB \350 \ \nDB geneB geneB GO:0006468 PMID:foo TAS P geneB \351 \ protein taxon:1000 20170308 DB \nDB geneB geneB GO:0005737\352 \ PMID:foo TAS C geneB protein taxon:1000 20170308 DB \353 \ \nDB geneB geneB GO:0004674 PMID:foo TAS F geneB \354 \ protein taxon:1000 20170308 DB \nDB geneB geneB GO:0006914\355 \ PMID:foo TAS P geneB protein taxon:1000 20170308 DB \356 \ \nDB geneB geneB GO:0016021 PMID:foo TAS C geneB \357 \ protein taxon:1000 20170308 DB \nDB geneB geneB GO:0015031\358 \ PMID:foo TAS P geneB protein taxon:1000 20170308 DB \359 \ \nDB geneA geneA GO:0006950 PMID:foo TAS P geneA \360 \ protein taxon:1000 20170308 DB \nDB geneA geneA GO:0005737\361 \ PMID:foo TAS C geneA protein taxon:1000 20170308 DB \362 \ \nDB geneC geneC GO:0003779 PMID:foo TAS F geneC \363 \ protein taxon:1000 20170308 DB \nDB geneC geneC GO:0006941\364 \ PMID:foo TAS P geneC protein taxon:1000 20170308 DB \365 \ \nDB geneC geneC GO:0005524 PMID:foo TAS F geneC \366 \ protein taxon:1000 20170308 DB \nDB geneC geneC GO:0003774\367 \ PMID:foo TAS F geneC protein taxon:1000 20170308 DB \368 \ \nDB geneC geneC GO:0005516 PMID:foo TAS F geneC \369 \ protein taxon:1000 20170308 DB \nDB geneC geneC GO:0005737\370 \ PMID:foo TAS C geneC protein taxon:1000 20170308 DB \371 \ \nDB geneC geneC GO:0005863 PMID:foo TAS C geneC \372 \ protein taxon:1000 20170308 DB \n\nNote that this is very much a pseudo-GAF\373 \ file since most of the fields apart from the gene name, GO term and sub-ontology\374 \ are fake. It should still work for what you need, however."375 - 'Question: As a small introductory project, I want to compare genome sequences376 of different strains of influenza virus.377 378 What are the publicly available databases of influenza virus gene/genome sequences?379 380 381 Answer: There area few different influenza virus database resources:382 383 384 The Influenza Research Database (IRD) (a.k.a FluDB - based upon URL)385 386 387 A NIAID Bioinformatics Resource Center or BRC which highly curates the data brought388 in and integrates it with numerous other relevant data types389 390 391 The NCBI Influenza Virus Resource392 393 394 A sub-project of the NCBI with data curated over and above the GenBank data that395 is part of the NCBI396 397 398 The GISAID EpiFlu Database399 400 401 A database of sequences from the Global Initiative on Sharing All Influenza Data.402 Has unique data from many countries but requires user agree to a data sharing403 policy.404 405 406 The OpenFluDB407 408 409 Former GISAID database that contains some sequence data that GenBank does not410 have.411 412 413 For those who also may be interested in other virus databases, there are:414 415 416 Virus Pathogen Resource (VIPR)417 418 419 A companion portal to the IRD, which hosts curated and integrated data for most420 other NIAID A-C virus pathogens including (but not limited to) Ebola, Zika, Dengue,421 Enterovirus, and Hepatitis C422 423 424 LANL HIV database425 426 427 Los Alamos National Laboratory HIV database with HIV data and many useful tools428 for all virus bioinformatics429 430 431 PaVE: Papilloma virus genome database (from quintik comment)432 433 434 NIAID developed and maintained Papilloma virus bioinformatics portal435 436 437 Disclaimer: I used to work for the IRD / VIPR and currently work for NIAID.'438 - "Question: I have a set of genomic ranges that are potentially overlapping. I\439 \ want to count the amount of ranges at certain positions using R. \nI'm Pretty\440 \ sure there are good solutions, but I seem to be unable to find them. \nSolutions\441 \ like cut or findIntervals don't achieve what I want as they only count on one\442 \ vector or accumulate by all values <= break.\nAlso countMatches {GenomicRanges}\443 \ doesn't seem to cover it.\nProbably one could use Bedtools, but I don't want\444 \ to leave R.\nI could only come up with a hilariously slow solution\n# generate\445 \ test data\ntestdata <- data.frame(chrom = rep(seq(1,10),10),\n \446 \ starts = abs(rnorm(100, mean = 1, sd = 1)) * 1000,\n \447 \ ends = abs(rnorm(100, mean = 2, sd = 1)) * 2000)\n\n# make sure that\448 \ all end coordinates are bigger than start\n# this is a requirement of the original\449 \ data\ntestdata <- testdata[testdata$ends - testdata$starts > 0,]\n\n# count\450 \ overlapping ranges on certain positions\ncount.data <- lapply(unique(testdata$chrom),\451 \ function(chromosome){\n tmp.inner <- lapply(seq(1,10000, by = 120), function(i){\n\452 \ sum(testdata$chrom == chromosome & testdata$starts <= i & testdata$ends\453 \ >= i)\n })\n return(unlist(tmp.inner))\n})\n\n# generate a data.frame\454 \ containing all data\ndf.count.data <- ldply(count.data, rbind)\n\n# ideally\455 \ the chromosome will be columns and not rows\nt(df.count.data)\n\nAnswer: GenomicRanges::countOverlaps\456 \ seems to be what you’re after:\nposition_range = GRanges(position$chrom, IRanges(position,\457 \ position, width = 1))\nranges_at_position = countOverlaps(position_ranges, granges)"458- source_sentence: samtools depth print out all positions459 sentences:460 - "Question: I have around ~3,000 short sequences of approximately ~10Kb long. What\461 \ are the best ways to find the motifs among all of these sequences? Is there\462 \ a certain software/method recommended?\nThere are several ways to do this. My\463 \ goal would be to:\n(1) Check for motifs repeated within individual sequences\n\464 (2) Check for motifs shared among all sequences\n(3) Check for the presence of\465 \ \"expected\" or known motifs\nWith respect to #3, I'm also curious if I find\466 \ e.g. trinucleotide sequences, how does one check the context around these regions?\n\467 Thank you for the recommendations/help!\n\nAnswer: For (3), this page has a lot\468 \ of links to pattern/motif finding tools. Following through the YMF link on that\469 \ page, I came across the University of Washington Motif Discovery section. Of\470 \ these projection seemed to be the only downloadable tool. I find it interesting\471 \ how old all these tools are; maybe the introduction of microarrays and NGS has\472 \ made them all redundant.\nYour sub-problem (2) seems similar to the problem\473 \ I'm having with Nippostrongylus brasiliensis genome sequences, where I'd like\474 \ to find regions of very high homology (length 500bp to 20kb or more, 95-99%\475 \ similar) that are repeated throughout the genome. These sequences are killing\476 \ the assembly.\nThe main way I can find these regions is by looking at a coverage\477 \ plot of long nanopore reads mapped to the assembled genome (using GraphMap or\478 \ BWA). Any regions with substantially higher than median coverage are likely\479 \ to be shared repeats.\nI've played around in the past with chopping up the reads\480 \ to smaller sizes, which works better for hitting smaller repeated regions that\481 \ are such a small proportion of most reads that they are never mapped to all\482 \ the repeated locations. I wrote my own script a while back to chop up reads\483 \ (for a different purpose), which produces a FASTA/FASTQ file where all reads\484 \ are exactly the same length. For some unknown reason I took the time to document\485 \ that script \"properly\" using POD, so here's a short summary:\n\nConverts all\486 \ sequences in the input FASTA file to the same length.\n Sequences shorter\487 \ than the target length are dropped, and sequences longer\n than the target\488 \ length are split into overlapping subsequences covering\n the entire range.\489 \ This prepares the sequences for use in an\n overlap-consensus assembler\490 \ requiring constant-length sequences (such as\n edena).\n\nAnd here's the\491 \ syntax:\n$ ./normalise_seqlengths.pl -h\nUsage:\n ./normalise_seqlengths.pl\492 \ <reads.fa> [options]\n\n Options:\n -help\n Only display this help\493 \ message\n\n -fraglength\n Target fragment length (in base-pairs, default\494 \ 2000)\n\n -overlap\n Minimum overlap length (in base-pairs, default\495 \ 200)\n\n -short\n Keep short sequences (shorter than fraglength)"496 - 'Question: Without going into too much background, I just joined up with a lab497 as a bioinformatics intern while I''m completing my masters degree in the field.498 The lab has data from an RNA-seq they outsourced, but the only problem is that499 the only data they have is preprocessed from the company that did the sequencing:500 filtering the reads, aligning them, and putting the aligned reads through RSEM.501 I currently have output from RSEM for each of the four samples consisting of:502 gene id, transcript id(s), length, expected count, and FPKM. I am attempting to503 get the FASTQ files from the sequencing, but for now, this is what I have, and504 I''m trying to get something out of it if possible.505 506 I found this article that talks about how expected read counts can be better than507 raw read counts when analyzing differential expression using EBSeq; it''s just508 one guy''s opinion, and it''s from 2014, so it may be wrong or outdated, but I509 thought I''d give it a try since I have the expected counts.510 511 However, I have just a couple of questions about running EBSeq that I can''t find512 the answers to:513 514 1: In the output RSEM files I have, not all genes are represented in each, about515 80% of them are, but for the ones that aren''t, should I remove them before analysis516 with EBSeq? It runs when I do, but I''m not sure if it is correct.517 518 2: How do I know which normalization factor to use when running EBSeq? This is519 more of a conceptual question rather than a technical question.520 521 Thanks!522 523 524 Answer: Yes, that blog post does represent just one guy''s opinion (hi!) and it525 does date all the way back to 2014, which is, like, decades in genomics years.526 :-) By the way, there is quite a bit of literature discussing the improvements527 that expected read counts derived from an Expectation Maximization algorithm provide528 over raw read counts. I''d suggest reading the RSEM papers for a start[1][2].529 530 But your main question is about the mechanics of running RSEM and EBSeq. First,531 RSEM was written explicitly to be compatible with EBSeq, so I''d be very surprised532 if it does not work correctly out-of-the-box. Second, EBSeq''s MedianNorm function533 worked very well in my experience for normalizing the library counts. Along those534 lines, the blog you mentioned above has another post that you may find useful.535 536 But all joking aside, these tools are indeed dated. Alignment-free RNA-Seq tools537 provide orders-of-magnitude improvements in runtime over the older alignment-based538 alternatives, with comparable accuracy. Sailfish was the first in a growing list539 of tools that now includes Salmon and Kallisto. When starting a new analysis from540 scratch (i.e. if you ever get the original FASTQ files), there''s really no good541 reason not to estimate expression using these much faster tools, followed by a542 differential expression analysis with DESeq2, edgeR, or sleuth.543 544 545 1Li B, Ruotti V, Stewart RM, Thomson JA, Dewey CN (2010) RNA-Seq gene expression546 estimation with read mapping uncertainty. Bioinformatics, 26(4):493–500, doi:10.1093/bioinformatics/btp692.547 548 2Li B, Dewey C (2011) RSEM: accurate transcript quantification from RNA-Seq data549 with or without a reference genome. BMC Bioinformatics, 12:323, doi:10.1186/1471-2105-12-323.'550 - 'Question: I am trying to use samtools depth (v1.4) with the -a option and a bed551 file listing the human chromosomes chr1-chr22, chrX, chrY, and chrM to print out552 the coverage at every position:553 554 cat GRCh38.karyo.bed | awk ''{print $3}'' | datamash sum 1555 556 3088286401557 558 559 I would like to know how to run samtools depth so that it produces 3,088,286,401560 entries when run against a GRCh38 bam file:561 562 samtools depth -b $bedfile -a $inputfile563 564 565 I tried it for a few bam files that were aligned the same way, and I get differing566 number of entries:567 568 3087003274569 570 3087005666571 572 3087007158573 574 3087009435575 576 3087009439577 578 3087009621579 580 3087009818581 582 3087010065583 584 3087010408585 586 3087010477587 588 3087010481589 590 3087012115591 592 3087013147593 594 3087013186595 596 3087013500597 598 3087149616599 600 601 Is there a special flag in samtools depth so that it reports all entries from602 the bed file?603 604 If samtools depth is not the best tool for this, what would be the equivalent605 with sambamba depth base?606 607 sambamba depth base --min-coverage=0 --regions $bedfile $inputfile608 609 610 Any other options?611 612 613 Answer: You might try using bedtools genomecov instead. If you provide the -d614 option, it reports the coverage at every position in the BAM file.615 616 bedtools genomecov -d -ibam $inputfile > "${inputfile}.genomecov"617 618 619 You can also provide a BED file if you just want to calculate in the target region.'620pipeline_tag: sentence-similarity621library_name: sentence-transformers622---623 624# SentenceTransformer based on BAAI/bge-small-en-v1.5625 626This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.627 628## Model Details629 630### Model Description631- **Model Type:** Sentence Transformer632- **Base model:** [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) <!-- at revision 5c38ec7c405ec4b44b94cc5a9bb96e735b38267a -->633- **Maximum Sequence Length:** 512 tokens634- **Output Dimensionality:** 384 dimensions635- **Similarity Function:** Cosine Similarity636<!-- - **Training Dataset:** Unknown -->637<!-- - **Language:** Unknown -->638<!-- - **License:** Unknown -->639 640### Model Sources641 642- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)643- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)644- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)645 646### Full Model Architecture647 648```649SentenceTransformer(650 (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel 651 (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})652 (2): Normalize()653)654```655 656## Usage657 658### Direct Usage (Sentence Transformers)659 660First install the Sentence Transformers library:661 662```bash663pip install -U sentence-transformers664```665 666Then you can load this model and run inference.667```python668from sentence_transformers import SentenceTransformer669 670# Download from the 🤗 Hub671model = SentenceTransformer("sentence_transformers_model_id")672# Run inference673sentences = [674 'samtools depth print out all positions',675 'Question: I am trying to use samtools depth (v1.4) with the -a option and a bed file listing the human chromosomes chr1-chr22, chrX, chrY, and chrM to print out the coverage at every position:\ncat GRCh38.karyo.bed | awk \'{print $3}\' | datamash sum 1\n3088286401\n\nI would like to know how to run samtools depth so that it produces 3,088,286,401 entries when run against a GRCh38 bam file:\nsamtools depth -b $bedfile -a $inputfile\n\nI tried it for a few bam files that were aligned the same way, and I get differing number of entries:\n3087003274\n3087005666\n3087007158\n3087009435\n3087009439\n3087009621\n3087009818\n3087010065\n3087010408\n3087010477\n3087010481\n3087012115\n3087013147\n3087013186\n3087013500\n3087149616\n\nIs there a special flag in samtools depth so that it reports all entries from the bed file?\nIf samtools depth is not the best tool for this, what would be the equivalent with sambamba depth base?\nsambamba depth base --min-coverage=0 --regions $bedfile $inputfile\n\nAny other options?\n\nAnswer: You might try using bedtools genomecov instead. If you provide the -d option, it reports the coverage at every position in the BAM file.\nbedtools genomecov -d -ibam $inputfile > "${inputfile}.genomecov"\n\nYou can also provide a BED file if you just want to calculate in the target region.',676 "Question: Without going into too much background, I just joined up with a lab as a bioinformatics intern while I'm completing my masters degree in the field. The lab has data from an RNA-seq they outsourced, but the only problem is that the only data they have is preprocessed from the company that did the sequencing: filtering the reads, aligning them, and putting the aligned reads through RSEM. I currently have output from RSEM for each of the four samples consisting of: gene id, transcript id(s), length, expected count, and FPKM. I am attempting to get the FASTQ files from the sequencing, but for now, this is what I have, and I'm trying to get something out of it if possible.\nI found this article that talks about how expected read counts can be better than raw read counts when analyzing differential expression using EBSeq; it's just one guy's opinion, and it's from 2014, so it may be wrong or outdated, but I thought I'd give it a try since I have the expected counts.\nHowever, I have just a couple of questions about running EBSeq that I can't find the answers to:\n1: In the output RSEM files I have, not all genes are represented in each, about 80% of them are, but for the ones that aren't, should I remove them before analysis with EBSeq? It runs when I do, but I'm not sure if it is correct.\n2: How do I know which normalization factor to use when running EBSeq? This is more of a conceptual question rather than a technical question.\nThanks!\n\nAnswer: Yes, that blog post does represent just one guy's opinion (hi!) and it does date all the way back to 2014, which is, like, decades in genomics years. :-) By the way, there is quite a bit of literature discussing the improvements that expected read counts derived from an Expectation Maximization algorithm provide over raw read counts. I'd suggest reading the RSEM papers for a start[1][2].\nBut your main question is about the mechanics of running RSEM and EBSeq. First, RSEM was written explicitly to be compatible with EBSeq, so I'd be very surprised if it does not work correctly out-of-the-box. Second, EBSeq's MedianNorm function worked very well in my experience for normalizing the library counts. Along those lines, the blog you mentioned above has another post that you may find useful.\nBut all joking aside, these tools are indeed dated. Alignment-free RNA-Seq tools provide orders-of-magnitude improvements in runtime over the older alignment-based alternatives, with comparable accuracy. Sailfish was the first in a growing list of tools that now includes Salmon and Kallisto. When starting a new analysis from scratch (i.e. if you ever get the original FASTQ files), there's really no good reason not to estimate expression using these much faster tools, followed by a differential expression analysis with DESeq2, edgeR, or sleuth.\n\n1Li B, Ruotti V, Stewart RM, Thomson JA, Dewey CN (2010) RNA-Seq gene expression estimation with read mapping uncertainty. Bioinformatics, 26(4):493–500, doi:10.1093/bioinformatics/btp692.\n2Li B, Dewey C (2011) RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics, 12:323, doi:10.1186/1471-2105-12-323.",677]678embeddings = model.encode(sentences)679print(embeddings.shape)680# [3, 384]681 682# Get the similarity scores for the embeddings683similarities = model.similarity(embeddings, embeddings)684print(similarities.shape)685# [3, 3]686```687 688<!--689### Direct Usage (Transformers)690 691<details><summary>Click to see the direct usage in Transformers</summary>692 693</details>694-->695 696<!--697### Downstream Usage (Sentence Transformers)698 699You can finetune this model on your own dataset.700 701<details><summary>Click to expand</summary>702 703</details>704-->705 706<!--707### Out-of-Scope Use708 709*List how the model may foreseeably be misused and address what users ought not to do with the model.*710-->711 712<!--713## Bias, Risks and Limitations714 715*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*716-->717 718<!--719### Recommendations720 721*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*722-->723 724## Training Details725 726### Training Dataset727 728#### Unnamed Dataset729 730* Size: 96 training samples731* Columns: <code>sentence_0</code> and <code>sentence_1</code>732* Approximate statistics based on the first 96 samples:733 | | sentence_0 | sentence_1 |734 |:--------|:----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|735 | type | string | string |736 | details | <ul><li>min: 6 tokens</li><li>mean: 14.93 tokens</li><li>max: 34 tokens</li></ul> | <ul><li>min: 103 tokens</li><li>mean: 397.92 tokens</li><li>max: 512 tokens</li></ul> |737* Samples:738 | sentence_0 | sentence_1 |739 |:-------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|740 | <code>Using shells other than bash</code> | <code>Question: As someone who's beginning to delve into bioinformatics, I'm noticing that like biology there are industry standards here, similar to Illumina in genomics and bowtie for alignment, many people use bash as shell. <br>Is using a shell besides bash going to cause issues for me?<br><br>Answer: Bioinformatics tools written in shell and other shell scripts generally specify the shell they want to use (via #!/bin/sh or e.g. #!/bin/bash if it matters), so won't be affected by your choice of user shell.<br>If you are writing significant shell scripts yourself, there are reasons to do it in a Bourne-style shell. See Csh Programming Considered Harmful and other essays/polemics.<br>A Bourne-style shell is pretty much the industry standard, and if you choose a substantially different shell you'll have to translate some of the documentation of your bioinformatics tools. It's not uncommon to have things like<br><br>Set some variables pointing at reference data and add the script to your PATH to run it:<br>export...</code> |741 | <code>Linear models of complex diseases</code> | <code>Question: A popular framework to analyze differences between groups, either experiments or diseases, in transcriptomics is using linear models (limma is a popular choice). <br>For instance we have a disease D with three stages as defined by clinicians, A, B and C. 10 samples each stage and the healthy H to compare with is RNA-sequenced. A typical linear model would be to observe the three stages~A+B+C independently. The data of each stage is not from the same person. (but for the question assume it isn't)<br>My understanding is that such a model would not take into account that stage C appears only on 30% of patients in stage B. And that a healthy patient upon external factors can jump to stage B. <br>If we want to find the role of a gene in the disease we should include somehow this information in the model. Which makes me think about mixing linear models and hidden Markov chains.<br>How can such a disease be described in terms of linear models with such data and information?<br><br>Answer: There are t...</code> |742 | <code>Detecting portions of human proteins with high degree of microbial similarity</code> | <code>Question: I'm a newcomer to the world of bioinformatics, and in need of help solving a problem.<br>My goal is to take a list of human proteins, and identify segments (13-17aa in length) with a high degree of similarity to microbial sequences. Ideally, I would like to start with list of FASTA sequences, and have an easy way to generate an output of the corresponding high similarity segments of each protein.<br>Are there existing tools or software that I should be aware of that will make my life easier?<br>Thanks in advance.<br><br>Answer: Sounds like precisely the job BLAST was developed for. Now, which flavor will depend on what you want to do and what data you have available. Some options:<br><br>PSI-BLAST: this is usually the best choice if you are trying to find protein homologs. It works by building a hidden markov model describing your query sequence and using that model to query a database of proteins. The advantage is that it is run in multiple iterations, giving you the chance to add or remove resu...</code> |743* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:744 ```json745 {746 "scale": 20.0,747 "similarity_fct": "cos_sim"748 }749 ```750 751### Training Hyperparameters752#### Non-Default Hyperparameters753 754- `per_device_train_batch_size`: 32755- `per_device_eval_batch_size`: 32756- `num_train_epochs`: 1757- `fp16`: True758- `batch_sampler`: no_duplicates759- `multi_dataset_batch_sampler`: round_robin760 761#### All Hyperparameters762<details><summary>Click to expand</summary>763 764- `overwrite_output_dir`: False765- `do_predict`: False766- `eval_strategy`: no767- `prediction_loss_only`: True768- `per_device_train_batch_size`: 32769- `per_device_eval_batch_size`: 32770- `per_gpu_train_batch_size`: None771- `per_gpu_eval_batch_size`: None772- `gradient_accumulation_steps`: 1773- `eval_accumulation_steps`: None774- `torch_empty_cache_steps`: None775- `learning_rate`: 5e-05776- `weight_decay`: 0.0777- `adam_beta1`: 0.9778- `adam_beta2`: 0.999779- `adam_epsilon`: 1e-08780- `max_grad_norm`: 1781- `num_train_epochs`: 1782- `max_steps`: -1783- `lr_scheduler_type`: linear784- `lr_scheduler_kwargs`: {}785- `warmup_ratio`: 0.0786- `warmup_steps`: 0787- `log_level`: passive788- `log_level_replica`: warning789- `log_on_each_node`: True790- `logging_nan_inf_filter`: True791- `save_safetensors`: True792- `save_on_each_node`: False793- `save_only_model`: False794- `restore_callback_states_from_checkpoint`: False795- `no_cuda`: False796- `use_cpu`: False797- `use_mps_device`: False798- `seed`: 42799- `data_seed`: None800- `jit_mode_eval`: False801- `use_ipex`: False802- `bf16`: False803- `fp16`: True804- `fp16_opt_level`: O1805- `half_precision_backend`: auto806- `bf16_full_eval`: False807- `fp16_full_eval`: False808- `tf32`: None809- `local_rank`: 0810- `ddp_backend`: None811- `tpu_num_cores`: None812- `tpu_metrics_debug`: False813- `debug`: []814- `dataloader_drop_last`: False815- `dataloader_num_workers`: 0816- `dataloader_prefetch_factor`: None817- `past_index`: -1818- `disable_tqdm`: False819- `remove_unused_columns`: True820- `label_names`: None821- `load_best_model_at_end`: False822- `ignore_data_skip`: False823- `fsdp`: []824- `fsdp_min_num_params`: 0825- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}826- `tp_size`: 0827- `fsdp_transformer_layer_cls_to_wrap`: None828- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}829- `deepspeed`: None830- `label_smoothing_factor`: 0.0831- `optim`: adamw_torch832- `optim_args`: None833- `adafactor`: False834- `group_by_length`: False835- `length_column_name`: length836- `ddp_find_unused_parameters`: None837- `ddp_bucket_cap_mb`: None838- `ddp_broadcast_buffers`: False839- `dataloader_pin_memory`: True840- `dataloader_persistent_workers`: False841- `skip_memory_metrics`: True842- `use_legacy_prediction_loop`: False843- `push_to_hub`: False844- `resume_from_checkpoint`: None845- `hub_model_id`: None846- `hub_strategy`: every_save847- `hub_private_repo`: None848- `hub_always_push`: False849- `gradient_checkpointing`: False850- `gradient_checkpointing_kwargs`: None851- `include_inputs_for_metrics`: False852- `include_for_metrics`: []853- `eval_do_concat_batches`: True854- `fp16_backend`: auto855- `push_to_hub_model_id`: None856- `push_to_hub_organization`: None857- `mp_parameters`: 858- `auto_find_batch_size`: False859- `full_determinism`: False860- `torchdynamo`: None861- `ray_scope`: last862- `ddp_timeout`: 1800863- `torch_compile`: False864- `torch_compile_backend`: None865- `torch_compile_mode`: None866- `include_tokens_per_second`: False867- `include_num_input_tokens_seen`: False868- `neftune_noise_alpha`: None869- `optim_target_modules`: None870- `batch_eval_metrics`: False871- `eval_on_start`: False872- `use_liger_kernel`: False873- `eval_use_gather_object`: False874- `average_tokens_across_devices`: False875- `prompts`: None876- `batch_sampler`: no_duplicates877- `multi_dataset_batch_sampler`: round_robin878 879</details>880 881### Framework Versions882- Python: 3.12.8883- Sentence Transformers: 3.4.1884- Transformers: 4.51.3885- PyTorch: 2.5.1+cu124886- Accelerate: 1.7.0887- Datasets: 3.2.0888- Tokenizers: 0.21.0889 890## Citation891 892### BibTeX893 894#### Sentence Transformers895```bibtex896@inproceedings{reimers-2019-sentence-bert,897 title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",898 author = "Reimers, Nils and Gurevych, Iryna",899 booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",900 month = "11",901 year = "2019",902 publisher = "Association for Computational Linguistics",903 url = "https://arxiv.org/abs/1908.10084",904}905```906 907#### MultipleNegativesRankingLoss908```bibtex909@misc{henderson2017efficient,910 title={Efficient Natural Language Response Suggestion for Smart Reply},911 author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},912 year={2017},913 eprint={1705.00652},914 archivePrefix={arXiv},915 primaryClass={cs.CL}916}917```918 919<!--920## Glossary921 922*Clearly define terms in order to be accessible across audiences.*923-->924 925<!--926## Model Card Authors927 928*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*929-->930 931<!--932## Model Card Contact933 934*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*935-->