diff --git a/.gitignore b/.gitignore
index 1d21541..49c40e4 100644
--- a/.gitignore
+++ b/.gitignore
@@ -2,5 +2,7 @@ nextflow
 .nextflow
 .nextflow.*
 pipeline_trace*
-nf_pipeline
-work
\ No newline at end of file
+nf_pipeline*
+work
+*.fa
+*.gtf
\ No newline at end of file
diff --git a/cleanup.sh b/cleanup.sh
new file mode 100755
index 0000000..24aa7bc
--- /dev/null
+++ b/cleanup.sh
@@ -0,0 +1,7 @@
+#!/bin/bash
+
+./nextflow main.nf -profile test
+rm -rf work
+rm -rf nf_pipeline
+rm .nextflow.log*
+rm pipeline_trace.txt*
\ No newline at end of file
diff --git a/conda.yml b/conda.yml
index 81cc47a..457fddc 100644
--- a/conda.yml
+++ b/conda.yml
@@ -6,7 +6,7 @@ channels:
   - conda-forge
 dependencies:
   # Default bismark
-  - salmon
-  - alevin-fry
-  - pyroe
+  - bioconda::salmon>=1.8.0
+  - bioconda::alevin-fry>=0.6.0
+  - bioconda::pyroe
   - bedtools
diff --git a/fastq.tar b/fastq.tar
new file mode 100644
index 0000000..1edb7bd
Binary files /dev/null and b/fastq.tar differ
diff --git a/genomic_data/fasta/genome.fa.gz b/genomic_data/fasta/genome.fa.gz
new file mode 100644
index 0000000..3ac8bb7
Binary files /dev/null and b/genomic_data/fasta/genome.fa.gz differ
diff --git a/input_files/pl_sheet_toy.tsv b/input_files/pl_sheet_toy.tsv
index 910d3af..120bde1 100644
--- a/input_files/pl_sheet_toy.tsv
+++ b/input_files/pl_sheet_toy.tsv
@@ -1,2 +1,2 @@
-reference	link
+chemistry	link
 v3	https://raw.githubusercontent.com/10XGenomics/cellranger/master/lib/python/cellranger/barcodes/3M-february-2018.txt.gz
diff --git a/input_files/ref_sheet.tsv b/input_files/ref_sheet.tsv
index 2a2f2f7..b83aa0d 100644
--- a/input_files/ref_sheet.tsv
+++ b/input_files/ref_sheet.tsv
@@ -1,3 +1,3 @@
 reference	link
 human2020A	https://cf.10xgenomics.com/supp/cell-exp/refdata-gex-GRCh38-2020-A.tar.gz
-mm10-2020A	https://cf.10xgenomics.com/supp/cell-exp/refdata-gex-mm10-2020-A.tar.gz
+mm10-2020A	https://cf.10xgenomics.com/supp/cell-exp/refdata-gex-mm10-2020-A.tar.gz
\ No newline at end of file
diff --git a/input_files/ref_sheet_toy.tsv b/input_files/ref_sheet_toy.tsv
index 7768c13..f99821d 100644
--- a/input_files/ref_sheet_toy.tsv
+++ b/input_files/ref_sheet_toy.tsv
@@ -1,2 +1,2 @@
-reference	link
-human-toy	https://umd.box.com/shared/static/7v6mof9emgudmuczc4lnwz4oou7xayks.gz
\ No newline at end of file
+reference	ref_data	fasta	gtf
+human-toy-download-ref	randommeaninglessfillerstring	genomic_data/fasta/genome.fa	genomic_data/genes/genes.gtf
diff --git a/input_files/sample_sheet_toy.tsv b/input_files/sample_sheet_toy.tsv
index ac15f40..2f0a034 100644
--- a/input_files/sample_sheet_toy.tsv
+++ b/input_files/sample_sheet_toy.tsv
@@ -1,2 +1,2 @@
 chemistry	reference	dataset_name	dataset_url	fastq_url	fastq_MD5sum	delete_fastq	feature_barcode_csv_url	multiplexing_library_csv_url
-v3	human-toy	toy set from 200 Sorted Cells from Human Glioblastoma Multiforme, 3' LT v3.1	https://www.10xgenomics.com/resources/datasets/200-sorted-cells-from-human-glioblastoma-multiforme-3-lt-v-3-1-3-1-low-6-0-0	https://umd.box.com/shared/static/bauvay4eq0v8xmu9e0fp0me7zj1qojnr.tar	8db8f1f92f92de0295efc9bb602a8276	1		
+v3	human-toy-download-ref	toy set from 200 Sorted Cells from Human Glioblastoma Multiforme, 3' LT v3.1	https://www.10xgenomics.com/resources/datasets/200-sorted-cells-from-human-glioblastoma-multiforme-3-lt-v-3-1-3-1-low-6-0-0	https://umd.box.com/shared/static/bauvay4eq0v8xmu9e0fp0me7zj1qojnr.tar	8db8f1f92f92de0295efc9bb602a8276	1		
\ No newline at end of file
diff --git a/main.nf b/main.nf
index 98f6832..2efa5a8 100755
--- a/main.nf
+++ b/main.nf
@@ -25,7 +25,6 @@ workflow {
       // producing both the knee filtered and 
       // unfiltered output
       preprocess()
-
   // merge permit list
       data = preprocess.out.chem_pl.cross(data)
             .map(it -> tuple(it[1][1], // reference
@@ -39,7 +38,6 @@ workflow {
                               it[1][7], // feature_barcode_csv_url
                               it[1][8]  // multiplexing_library_csv_url
             ))
-
   // merge t2g and salmon index
       data = preprocess.out.ref_t2g_index.cross(data)
             .map(it -> tuple( it[1][1], // chemistry
@@ -55,7 +53,11 @@ workflow {
                               it[1][2], // pl_path
                               it[0][1] // t2g_path
             ))
-
       salmon_map(data) 
-      af(salmon_map.out, Channel.value("unfilt"))
+      if(!salmon_map.out[0].equals("v2") && !salmon_map.out[0].equals("v3")) {
+            af(salmon_map.out, Channel.value("unfilt"))
+      } else {
+            af(salmon_map.out, Channel.value("knee"))
+      }
+      af.out.view()
 }
diff --git a/modules/af.nf b/modules/af.nf
index 94764af..306357d 100644
--- a/modules/af.nf
+++ b/modules/af.nf
@@ -270,6 +270,10 @@ workflow af {
         | alevin_fry_collate \
         | alevin_fry_quant \
         | write_description
+
+    emit:
+        write_description.out
+
 }
 
 
diff --git a/modules/preprocess.nf b/modules/preprocess.nf
index 942d252..05f6e24 100644
--- a/modules/preprocess.nf
+++ b/modules/preprocess.nf
@@ -1,3 +1,4 @@
+import groovy.io.FileType
 
 /*
 * based on the provided reference and link of permit list, 
@@ -14,8 +15,7 @@ process get_permitlist {
         tuple val(chemistry), file(ofile_name), emit: chem_pl
 
     script:
-        ofile_name = link.getName()
-
+        ofile_name = link.getName()    
     if (link.getExtension() != 'txt') {
         ofile_name = link.getBaseName()
         """
@@ -32,49 +32,6 @@ process get_permitlist {
 // permitlist = permitlist.flatMap { ["${it[0]}" : it[1]] }
 
 
-
-/*
-* based on the provided reference and link of cellranger references, 
-* download and uncompress on the fly, and make splici for them 
-* NOTE: 
-* 1. require a tsv named ref_sheet.tsv in the ${bench_dir}/input_files
-* 2. only work for pre-built cellranger references, like humand2020A and mm10-2020A
-* 3. if other reference needed, this process has to be expanded to 
-*    enable cellranger to make reference 
-*/
-process get_splici {
-    tag "get_permitlist:${reference}"
-    // conda "bioconda::bedtools bioconda::pyroe"
-
-    input:
-        tuple val(reference), val(link)
-        
-    output:
-        tuple val(reference),
-        path("splici_$reference/splici_fl${params.read_len - 5}.fa"), 
-        path("splici_$reference/splici_fl${params.read_len - 5}_t2g_3col.tsv"), 
-        emit: splici
-
-    script:
-        """
-        mkdir reference_$reference && wget -qO- ${link} | tar xzf - --strip-components=1 -C reference_$reference
-        pyroe make-splici reference_$reference/fasta/genome.fa reference_$reference/genes/genes.gtf ${params.read_len} splici_$reference
-        rm -rf reference_$reference
-        """
-    stub:
-        """
-            mkdir reference_$reference
-            mkdir reference_$reference/genes && touch reference_$reference/genes/genes.gtf
-            mkdir reference_$reference/fasta && touch reference_$reference/fasta/genome.fa
-
-            mkdir splici_$reference
-            touch splici_$reference/splici_fl${params.read_len-5}.fa
-            touch splici_$reference/splici_fl${params.read_len-5}_t2g_3col.tsv
-            rm -rf reference_$reference
-        """
-}
-
-
 /*
 * This process takes an input sample
 * as defined by the columns in the sample table csv
@@ -122,6 +79,292 @@ process salmon_index {
         """
 }
 
+/*
+* Process sets up reference data based on the provided reference link or local path,
+* which could contain data for both gtf and fasta files, just one, or neither if user 
+* provides seperate paths or remote links to both files
+*/
+
+process process_ref_data {
+    tag "ref:${reference}"
+
+    input: 
+        tuple val(reference), val(ref_data), val(fasta), val(gtf)
+    output:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf),
+        emit: processed_ref_data        
+
+    script:
+        ref_path = "reference_${reference}/"
+    if (ref_data.length () > 3 && ref_data.substring(0,4).equals("http")) { //check if ref_data is https or local
+        """
+        mkdir $ref_path
+        wget -qO- $ref_data | tar xzf - --strip-components=1 -C $ref_path
+        """
+    } else if(ref_data.contains("gz") || ref_data.contains("tar")) { //if its local see if its a tarball or if its just fasta and gtf
+        """
+        mkdir $ref_path
+        tar xzf $ref_data --strip-components=1 -C $ref_path
+        """
+    } else {
+        ref_data_path = file(ref_data)
+        if(ref_data_path.exists()) {
+            """
+            mkdir $ref_path
+            cp -a $ref_data/. $ref_path
+            """
+        } else {
+            """
+            mkdir $ref_path
+            """
+        }
+        
+    }
+}
+
+/*
+* Process sets up gtf and fasta data based on the provided input in ref_sheet;
+* if it is not a link, standardizes it to a absolute path for future use
+*/
+
+process standardize_files {
+    tag "standardize_files:$reference"
+
+    input: 
+        tuple val(reference), path(ref_path), val(fasta), val(gtf)
+    output:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf),
+        emit: standardized_files_data
+    script:
+        if(fasta.length () > 3 && !fasta.substring(0,4).equals("http")) { //fasta is not a url, and is a file
+            if(!fasta.substring(0,1).equals("/")) {
+                fasta = "/" + "$fasta"
+            }
+            ref_fasta = file(ref_path.toRealPath() + "$fasta") //fasta is within given reference path
+            rel_fasta = file("${projectDir}${fasta}") //fasta is relative to project directory
+            abs_fasta = file("$fasta") //fasta is absolute path
+            if (ref_fasta.exists()) { //fasta exisits within given reference path
+                fasta = ref_fasta.toRealPath()
+            } else if (rel_fasta.exists()) { //fasta exists relative to project directory
+                fasta = rel_fasta.toRealPath()
+            } else if (abs_fasta.exists()) {
+                //fasta is an absolute path
+            }
+        }
+        if(gtf.length () > 3 && !gtf.substring(0,4).equals("http")) {   
+            if(!gtf.substring(0,1).equals("/")) {
+                gtf = "/" + "$gtf"
+            }
+            ref_gtf = file(ref_path.toRealPath() + "$gtf")
+            rel_gtf = file("${projectDir}${gtf}")
+            abs_gtf = file("$gtf")
+            if (ref_gtf.exists()) { //gtf exisits within given reference path
+                gtf = ref_gtf.toRealPath()
+            } else if (rel_gtf.exists()) { //gtf exists relative to project directory
+                gtf = rel_gtf.toRealPath()
+            } else if (abs_gtf.exists()) {
+                //gtf is an absolute path
+            }
+        }
+        """
+
+        """
+        
+}
+
+/*
+* Downloads, unzips, or moves provided fasta file in ref_sheet to 
+* correct location to be used in splici generation
+*/
+process standard_fasta {
+    tag "standard_fasta:$reference"
+
+    input:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf)
+    output:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf),
+        emit: processed_fasta_path    
+    script:
+        fasta = fasta.toString()
+        abs_fasta = file("$fasta")
+        target = ref_path.toRealPath() + "/fasta/genome.fa"
+        target = target.toString()
+        if (fasta.length () > 3 && fasta.substring(0,4).equals("http") && fasta.contains(".tar")) { //fasta is remote tarball
+            //download and untar
+            """
+            mkdir -p $ref_path/fasta/
+            wget -qO- $fasta | tar xzf - --strip-components=1 -C $ref_path/fasta/genome.fa
+            """
+        } else if (fasta.length () > 3 && fasta.substring(0,4).equals("http") && fasta.contains(".gz")) { //fasta is remote gzip
+            //download and unzip
+            """
+            mkdir -p $ref_path/fasta/
+            wget -qO- $fasta | gunzip -c > $ref_path/fasta/genome.fa
+            """
+        }  else if (fasta.length () > 3 && fasta.substring(0,4).equals("http")) { //fasta is remote
+            //download and move
+            """
+            mkdir -p $ref_path/fasta/
+            wget -qP $fasta $ref_path/fasta/genome.fa
+            """
+        } else if (abs_fasta.exists()) { //if the path given is an absolute path to a file
+            if (fasta.contains(".tar")) { //tarball 
+                //untar
+                """
+                mkdir -p $ref_path/fasta/
+                tar -xzf $fasta --strip-components=1 -C $ref_path/fasta/genome.fa
+                """
+            } else if (fasta.contains(".gz")) { //gzip
+                //gunzip
+                """
+                mkdir -p $ref_path/fasta/
+                gunzip $fasta -c > $ref_path/fasta/genome.fa
+                """
+            } else if(!fasta.equals(target)){ //.fa and its not already where it needs to be
+                //move file to location exisiting
+                """
+                mkdir -p $ref_path/fasta
+                cp $fasta $target
+                """
+            } else { //file exactly where it needs to be
+                """
+                # this means something is wrong with input
+                """
+            }
+        }
+}
+
+/*
+* Downloads, unzips, or moves provided gtf file in ref_sheet to 
+* correct location to be used in splici generation
+*/
+process standard_gtf {
+    tag "standard_gtf:$reference"
+
+    input:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf)
+    output:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf),
+        emit: processed_gtf_path    
+    script:
+        gtf = gtf.toString()
+        abs_gtf = file("$gtf")
+        target = ref_path.toRealPath() + "/genes/genes.gtf"
+        target = target.toString()
+        if (gtf.length() > 3 && gtf.substring(0,4).equals("http") && gtf.contains(".tar")) { //gtf is remote tarball
+            //download and untar
+            """
+            mkdir -p $ref_path/genes/
+            wget -qO- $gtf | tar xzf - --strip-components=1 -C $ref_path/genes/genes.gtf
+            """
+        } else if (gtf.length() > 3 && gtf.substring(0,4).equals("http") && gtf.contains(".gz")) { //gtf is remote gzip
+            //download and unzip
+            """
+            mkdir -p $ref_path/genes/
+            wget -qO- $gtf | gunzip -c > $ref_path/genes/genes.gtf
+            """
+        }  else if (gtf.length() > 3 && gtf.substring(0,4).equals("http")) { //gtf is remote
+            //download and move
+            """
+            mkdir -p $ref_path/genes
+            wget -qP $gtf $ref_path/genes/genes.gtf
+            """
+        } else if (abs_gtf.exists()) { //if the path given is an absolute path to a file
+            if (gtf.contains(".tar")) { //tarball 
+                //untar
+                """
+                mkdir -p $ref_path/genes/
+                tar -xzf $gtf --strip-components=1 -C $ref_path/genes/genes.gtf
+                """
+            } else if (gtf.contains(".gz")) { //gzip
+                //gunzip
+                """
+                mkdir -p $ref_path/genes/
+                gunzip $gtf -c > $ref_path/genes/genes.gtf
+                """
+            } else if(!gtf.equals(target)){ //.gtf and its not already where it needs to be
+                //move file to location exisiting
+                """
+                mkdir -p $ref_path/genes
+                cp $gtf $ref_path/genes/genes.gtf
+                """
+            } else { //file exactly where it needs to be
+                """
+
+                """
+            }
+        } else {
+            """
+            # this means something is wrong with input
+            """
+        }
+}
+
+process prep_commands {
+    tag "prep_commands:${reference}"
+
+    input:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf)
+    output:
+        tuple val(reference), path(ref_path), val(fasta), val(gtf), val(pyroe_command_args), val(file_name)
+    script:
+        pyroe_command_args = ""
+        file_name = "splici"
+        if (!params.pyroe.filename_pref.equals("")) {
+            file_name = params.pyroe.filename_pref
+            pyroe_command_args += " --filename-prefix $params.pyroe.filename_pref"
+        }
+        if (!params.pyroe.bt_path.equals("")) {
+            pyroe_command_args += " --bt-path $params.pyroe.bt_path"
+        }
+        if (params.pyroe.dedup_seqs.equalsIgnoreCase("True") || params.pyroe.dedup_seqs.equalsIgnoreCase("T")) {
+            pyroe_command_args += " --dedup-seqs"
+        }
+        if (params.pyroe.no_flanking_merge.equalsIgnoreCase("True") || params.pyroe.no_flanking_merge.equalsIgnoreCase("T")) {
+            pyroe_command_args += " --no-flanking-merge"
+        }
+        """
+
+        """
+}   
+
+/*
+* based on the provided and processed reference sheet, and make splici for them 
+* NOTE: 
+* 1. require a tsv named ref_sheet.tsv in the ${bench_dir}/input_files
+* 2. any combination of local and remote, zipped/unzipped should be accepted
+*/
+process get_splici {
+    tag "get_splici:${reference}"
+    // conda "bioconda::bedtools bioconda::pyroe"
+
+    input:
+        tuple val(reference), path(ref_path), path(fasta), path(gtf), val(args), val(file_name)
+        
+    output:
+        tuple val(reference),
+        path("splici_$reference/${file_name}_fl${params.pyroe.read_len - params.pyroe.trim_length}.fa"), 
+        path("splici_$reference/${file_name}_fl${params.pyroe.read_len - params.pyroe.trim_length}_t2g_3col.tsv"), 
+        emit: splici
+
+    script:
+        """
+        pyroe make-splici $ref_path/fasta/genome.fa $ref_path/genes/genes.gtf ${params.pyroe.read_len} splici_$reference${args}
+        rm -rf ${ref_path}/
+        """
+    stub:
+        """
+            mkdir reference_$reference
+            mkdir reference_$reference/genes && touch reference_$reference/genes/genes.gtf
+            mkdir reference_$reference/fasta && touch reference_$reference/fasta/genome.fa
+            mkdir splici_$reference
+            touch splici_$reference/${file_name}${params.pyroe.read_len - params.pyroe.trim_length}.fa
+            touch splici_$reference/${file_name}${params.pyroe.read_len - params.pyroe.trim_length}_t2g_3col.tsv
+            rm -rf reference_$reference
+        """
+}
+
+
 /*
 * This workflow take a ref_sheet.tsv and a pl_sheet.tsv file 
 * in the ${bench_dir}/input_files
@@ -134,23 +377,31 @@ workflow preprocess {
     pl_sheet = Channel
             .fromPath(params.input_sheets.permitlist)
             .splitCsv(header:true, sep:"\t", strip: true)
-            .map{ row-> tuple(row.reference,
+            .map{ row-> tuple(row.chemistry,
                             row.link)
             }
     get_permitlist(pl_sheet)
-
+    
     ref_sheet = Channel
                 .fromPath(params.input_sheets.reference)
                 .splitCsv(header:true, sep:"\t", strip: true)
                 .map{ row-> tuple(row.reference,
-                                row.link)
+                                row.ref_data, row.fasta, row.gtf)
                 }
-    get_splici(ref_sheet)
+    process_ref_data(ref_sheet)
+    standardize_files(process_ref_data.out)
+
+    standard_fasta(standardize_files.out)
+    standard_gtf(standard_fasta.out)
+    //prepare commands for pyroe based on user input in config sheet
+    prep_commands(standard_gtf.out)
+
+    get_splici(prep_commands.out) 
+    
     salmon_index(get_splici.out)
     
     chem_pl = get_permitlist.out.chem_pl
     ref_t2g_index = salmon_index.out.ref_t2g_index
-
     emit:
         chem_pl
         ref_t2g_index
diff --git a/modules/salmon_map.nf b/modules/salmon_map.nf
index 80a2e48..7e073bf 100644
--- a/modules/salmon_map.nf
+++ b/modules/salmon_map.nf
@@ -3,14 +3,14 @@ workflow salmon_map {
         samp 
 
     main: 
-        download_fastq(samp) | salmon_map_rad
+        standardize_salmon_files(samp) | download_fastq | salmon_map_rad
 
     emit:
         salmon_map_rad.out.chemistry
         salmon_map_rad.out.reference
         salmon_map_rad.out.dataset_name
         salmon_map_rad.out.dataset_url
-        salmon_map_rad.out.fastq_url
+        salmon_map_rad.out.fastq_file
         salmon_map_rad.out.fastq_MD5sum
         salmon_map_rad.out.delete_fastq
         salmon_map_rad.out.feature_barcode_csv_url
@@ -23,23 +23,16 @@ workflow salmon_map {
 
 }
 
-/*
-* This process takes an input sample
-* as defined by the columns in the sample table csv
-* and runs `alevin` on it to produce a RAD file.
-* Currently, sketch mode is always used.
-*/
-
-process download_fastq {
-    tag "download_fastq:${fastq_MD5sum}"
-    label 'single_threads'
-
+process standardize_salmon_files {
+    tag "standardize_salmon_files:${dataset_name}"
+    errorStrategy 'terminate'
+    
     input:
         tuple val(chemistry), 
             val(reference), 
             val(dataset_name), 
             val(dataset_url), 
-            val(fastq_url), 
+            val(fastq_file), 
             val(fastq_MD5sum), 
             val(delete_fastq), 
             val(feature_barcode_csv_url), 
@@ -47,13 +40,12 @@ process download_fastq {
             path(index_dir_path), 
             path(pl_path), 
             path(t2g_path)
-    
     output:
         val chemistry, emit: chemistry
         val reference, emit: reference
         val dataset_name, emit: dataset_name
         val dataset_url, emit: dataset_url
-        val fastq_url, emit:fastq_url
+        val fastq_file, emit:fastq_file
         val fastq_MD5sum, emit: fastq_MD5sum
         val delete_fastq, emit: delete_fastq
         val feature_barcode_csv_url, emit: feature_barcode_csv_url
@@ -61,28 +53,122 @@ process download_fastq {
         path index_dir_path, emit: index_dir_path
         path pl_path, emit: pl_path
         path t2g_path, emit: t2g_path
-        path "${fastq_MD5sum}_fastqs", emit: fastq_dir_path
 
-    """
-        num_attempt=1
-        wget ${fastq_url} -P ${fastq_MD5sum}_cfastqs
-        while [ "\$(md5sum ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) | cut -d' ' -f1)" != "${fastq_MD5sum}" ]
-        do
-            if [[ \$num_attempt -gt 3 ]]
-            then
-                echo "Three attempts were made to fetch ${fastq_url.toString().lastIndexOf('/').with {fastq_url.toString().substring(it+1, fastq_url.toString().length())}}, but the MD5sum (\$(md5sum ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) | cut -d' ' -f1)) of the downloaded file \$(ls ${fastq_MD5sum}_cfastqs) didn't match the expected MD5sum ($fastq_MD5sum). Processing of this dataset will not proceed. Please check the MD5sum and internet connectivity."
-                exit 1
-            fi
-            let "num_attempt+=1"
-            rm -rf ${fastq_MD5sum}_cfastqs
-            wget ${fastq_url} -P ${fastq_MD5sum}_cfastqs
-        done
-        mkdir -p ${fastq_MD5sum}_fastqs
-        tar xf ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) --strip-components=1 -C ${fastq_MD5sum}_fastqs
-        rm -rf ${fastq_MD5sum}_cfastqs
-    """
+    script:
+        //list of valid chem options, including chromiumv2 and v3 in case user types full name over v2 or v3
+        valid_chems = ["dropseq", "v2", "v3", "chromiumV3", "chromium", "gemcode", 
+        "citeseq", "celseq", "celseq2", "splitseqV1", "splitseqV2", 
+        "quartzseq2", "sciseq3"]
+        if(!valid_chems.contains(chemistry)) {
+            error "Invalid chemistry option: " + chemistry
+        }
+        if(fastq_file.length () > 3 && !fastq_file.substring(0,4).equals("http")) { //fastq is not a url, and is a file
+            if(!fastq_file.substring(0,1).equals("/")) {
+                fastq_file = "/" + "$fastq_file"
+            }
+            rel_fastq = file("${projectDir}${fastq_file}") //fastq is relative to project directory
+            abs_fastq = file("$fastq_file") //fastq is absolute path
+            if (rel_fastq.exists()) { //fastq exists relative to project directory
+                fastq_file = rel_fastq.toRealPath()
+            } else if (!abs_fastq.exists()) {
+                error "Could not find referenced fastq file."
+            }
+        }
+        //todo csvs for multiplex and feature barcodes?
+        """
+
+        """
+    
+}
+
+/*
+* This process takes an input sample
+* as defined by the columns in the sample table csv
+* and runs `alevin` on it to produce a RAD file.
+* Currently, sketch mode is always used.
+*/
+
+process download_fastq {
+    tag "download_fastq:${fastq_MD5sum}"
+    label 'single_threads'
 
+    input:
+        val chemistry 
+        val reference
+        val dataset_name 
+        val dataset_url 
+        val fastq_file 
+        val fastq_MD5sum 
+        val delete_fastq 
+        val feature_barcode_csv_url 
+        val multiplexing_library_csv_url
+        path index_dir_path
+        path pl_path
+        path t2g_path
 
+    output:
+        val chemistry, emit: chemistry
+        val reference, emit: reference
+        val dataset_name, emit: dataset_name
+        val dataset_url, emit: dataset_url
+        val fastq_file, emit:fastq_file
+        val fastq_MD5sum, emit: fastq_MD5sum
+        val delete_fastq, emit: delete_fastq
+        val feature_barcode_csv_url, emit: feature_barcode_csv_url
+        val multiplexing_library_csv_url, emit: multiplexing_library_csv_url
+        path index_dir_path, emit: index_dir_path
+        path pl_path, emit: pl_path
+        path t2g_path, emit: t2g_path
+        path "${fastq_MD5sum}_fastqs", emit: fastq_dir_path
+    
+    script:
+        fastq_file = fastq_file.toString()
+        abs_fastq = file("$fastq_file")
+        if(fastq_file.length() > 3 && fastq_file.substring(0,4).equals("http")) { //fastq is a URL
+            """
+                num_attempt=1
+                wget ${fastq_file} -P ${fastq_MD5sum}_cfastqs
+                while [ "\$(md5sum ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) | cut -d' ' -f1)" != "${fastq_MD5sum}" ]
+                do
+                    if [[ \$num_attempt -gt 3 ]]
+                    then
+                        echo "Three attempts were made to fetch ${fastq_file.toString().lastIndexOf('/').with {fastq_file.toString().substring(it+1, fastq_file.toString().length())}}, but the MD5sum (\$(md5sum ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) | cut -d' ' -f1)) of the downloaded file \$(ls ${fastq_MD5sum}_cfastqs) didn't match the expected MD5sum ($fastq_MD5sum). Processing of this dataset will not proceed. Please check the MD5sum and internet connectivity."
+                        exit 1
+                    fi
+                    let "num_attempt+=1"
+                    rm -rf ${fastq_MD5sum}_cfastqs
+                    wget ${fastq_file} -P ${fastq_MD5sum}_cfastqs
+                done
+                mkdir -p ${fastq_MD5sum}_fastqs
+                tar xf ${fastq_MD5sum}_cfastqs/\$(ls ${fastq_MD5sum}_cfastqs) --strip-components=1 -C ${fastq_MD5sum}_fastqs
+                rm -rf ${fastq_MD5sum}_cfastqs
+            """
+        } else if(abs_fastq.exists()) { //check that the fastq file exists locally
+            if (fastq_MD5sum.length() != 32) {
+                fastq_MD5sum = fastq_file.md5()
+            }
+            if(fastq_file.length () > 3 && fastq_file.contains(".tar") && fastq_file.contains(".gz")) {
+                """
+                mkdir ${fastq_MD5sum}_fastqs
+                tar -xzf ${fastq_file} --strip-components=1 -C ${fastq_MD5sum}_fastqs
+                """
+            } else if(fastq_file.length () > 3 && fastq_file.contains(".tar")) {
+                """
+                mkdir ${fastq_MD5sum}_fastqs
+                tar -xf ${fastq_file} --strip-components=1 -C ${fastq_MD5sum}_fastqs
+                """
+            } else if (fastq_file.length () > 3 && fastq_file.contains(".gz")) {
+                """
+                mkdir ${fastq_MD5sum}_fastqs
+                gunzip ${fastq_file} -c > ${fastq_MD5sum}_fastqs
+                """
+            } else {
+                """
+                mkdir ${fastq_MD5sum}_fastqs
+                cp -a ${fastq_file} ${fastq_MD5sum}_fastqs
+                """
+            }
+        }
 }
 
 process salmon_map_rad {
@@ -95,7 +181,7 @@ process salmon_map_rad {
         val reference
         val dataset_name
         val dataset_url
-        val fastq_url
+        val fastq_file
         val fastq_MD5sum
         val delete_fastq
         val feature_barcode_csv_url
@@ -110,7 +196,7 @@ process salmon_map_rad {
         val reference, emit: reference
         val dataset_name, emit: dataset_name
         val dataset_url, emit: dataset_url
-        val fastq_url, emit:fastq_url
+        val fastq_file, emit:fastq_file
         val fastq_MD5sum, emit: fastq_MD5sum
         val delete_fastq, emit: delete_fastq
         val feature_barcode_csv_url, emit: feature_barcode_csv_url
diff --git a/nextflow.config b/nextflow.config
index 0adc379..c213b4e 100755
--- a/nextflow.config
+++ b/nextflow.config
@@ -13,7 +13,6 @@ params {
     work_dir = "${projectDir}/nf_pipeline"
     output_dir = "${projectDir}/nf_pipeline/output"
     // delete_fastq = 1
-    read_len = 90
     n_threads = 16
 
     input_sheets {
@@ -23,6 +22,19 @@ params {
         reference = "${projectDir}/input_files/ref_sheet.tsv"
         sample = "${projectDir}/input_files/sample_sheet.tsv"
     }
+    pyroe {
+        read_len = 90
+        trim_length = 5
+
+        filename_pref = "new_file_name"
+        bt_path = ""
+        
+        //the following commands can be left as "", put True or T to run pyroe with that option
+
+        no_bt = "False"
+        dedup_seqs = "False"
+        no_flanking_merge = "False"
+    }
 }
 
 process {