diff --git a/CHANGELOG b/CHANGELOG
index 928f26c2..d55ffc64 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,16 @@
+v1.4.5:
+    * Fieldbioinformatics now supports rapid barcoded (fragmented) primer trimming and normalisation
+    * Nanopolish has been removed completely due to several compatibility issues
+    * Medaka has also been removed completely due to kicking out long indels in a way that cannot be changed.
+    * Clair3 is now the default variant caller, by default only the r9.4.1 models are available but a artic_get_models command has been added which will fetch the ONT created r10.4.1 models listed in the rerio repository.
+    * The pipeline will also attempt to pick an appropriate model based on the basecall_model_version_id field that is added to read headers by default by ONT sequencers.
+    * Removed longshot entirely, it also kicks out long variants and is now unnecessary due to clair3 being a much better variant caller.
+    * Primer scheme fetcher has been updated to pull from the quick-lab primal hub schemes repository. For schemes not available in this repository you may provide them directly with the arguments --bed and --ref.
+    * Automated docker builds pushing to quay.io for use in nextflow pipelines etc.
+    * Remove some old functionality which is no longer relevant (basecalling, gather, etc)
+    * Re-implement CI as a gh action.
+    * Fix the overlapping variants issue by normalising variants against the pre-consensus using bcftools norm.
+
 v1.1.0-rc1:
     * Support for read groups:
         * Support ‘pool’ read groups taken from BED file, e.g.:
diff --git a/README.md b/README.md
index a2d5cf9a..bb72296d 100644
--- a/README.md
+++ b/README.md
@@ -25,8 +25,6 @@ Features include:
 - variant calling
 - consensus building
 
-There are **2 workflows** baked into this pipeline, one which uses signal data (via [nanopolish](https://github.com/jts/nanopolish)) and one that does not (via [medaka](https://github.com/nanoporetech/medaka)).
-
 <!-- ## Installation
 
 ### Via conda
diff --git a/artic/pipeline.py b/artic/pipeline.py
index eb33425f..97852aa1 100755
--- a/artic/pipeline.py
+++ b/artic/pipeline.py
@@ -66,9 +66,6 @@ def init_pipeline_parser():
     parser_minion = subparsers.add_parser(
         "minion", help="Run the alignment/variant-call/consensus pipeline"
     )
-    # parser_minion.add_argument(
-    #     "scheme", metavar="scheme", help="The name of the scheme"
-    # )
     parser_minion.add_argument(
         "sample", metavar="sample", help="The name of the sample"
     )
@@ -79,7 +76,7 @@ def init_pipeline_parser():
     parser_minion.add_argument(
         "--model-path",
         metavar="model_path",
-        help="Path containing clair3 models, defaults to models packaged with conda installation",
+        help="Path containing clair3 models, defaults to models packaged with conda installation (default: $CONDA_PREFIX/bin/models/)",
         type=str,
     )
     parser_minion.add_argument(
diff --git a/artic/utils.py b/artic/utils.py
index d4dcd369..6a3fe198 100644
--- a/artic/utils.py
+++ b/artic/utils.py
@@ -784,7 +784,16 @@ def get_scheme_legacy(scheme_name, scheme_directory, scheme_version="1"):
     raise SystemExit(1)
 
 
-def choose_model(read_file: str):
+def choose_model(read_file: str) -> dict:
+    """
+    Choose the appropriate clair3 model based on the `basecall_model_version_id` field in the read header (if it exists)
+
+    Args:
+        read_file (str): Path to the fastq file
+
+    Returns:
+        dict: The chosen clair3 model as a dictionary
+    """
 
     models_class = clair3_manifest()
     models = models_class.models
diff --git a/docs/commands.md b/docs/commands.md
index ef1338c7..75fe19e5 100644
--- a/docs/commands.md
+++ b/docs/commands.md
@@ -5,127 +5,13 @@ authors:
   - Sam Wilkinson
   - Will Rowe
   - Nick Loman
-date: 2024-08-16
+date: 2024-11-11
 ---
 
 # Commands
 
 This page documents the available commands via the `artic` command line interface.
 
-## demultiplex
-
-### Overview
-
-Run demultiplex
-
-### Input
-
-- undemultiplexed FASTA file
-
-### Output
-
-- demultiplexed FASTA file(s)
-
-### Usage example
-
-```bash
-artic demultiplex <fasta>
-```
-
-| Argument name(s)      | Required | Default value | Description                    |
-| :-------------------- | :------- | :------------ | :----------------------------- |
-| fasta                 | Y        | NA            | The undemultiplexed FASTA file |
-| --threads             | N        | 8             | The number of threads          |
-| --prefix              | N        | NA            | Prefix for demultiplexed files |
-| --no-remove-directory | N        | NA            | Don't remove the directory     |
-
----
-
-## export
-
-### Overview
-
-The export command is used to make a redistributable package of data for re-analysis. This includes the FASTQ file, the sequencing summary and the FAST5 file. The selection of reads to be used comes from a BAM file, and only aligned reads are used.
-
-### Input
-
-- a completed minion pipeline run
-
-### Output
-
-- a redistributable package of data
-
-### Usage example
-
-```bash
-artic export <prefix> <bamfile> <sequencing_summary> <fast5_directory> <output_directory>
-```
-
-| Argument name(s)   | Required | Default value | Description                          |
-| :----------------- | :------- | :------------ | :----------------------------------- |
-| prefix             | Y        | NA            | The run prefix                       |
-| bamfile            | Y        | NA            | The BAM file to export reads from    |
-| sequencing_summary | Y        | NA            | Path to Guppy sequencing summary     |
-| fast5_directory    | Y        | NA            | The path to directory of FAST5 files |
-| output_directory   | Y        | NA            | The path to export the data to       |
-
----
-
-## extract
-
-### Overview
-
-Create an empty poredb database
-
-### Input
-
-- na
-
-### Output
-
-- an initialised poredb database
-
-### Usage example
-
-```bash
-artic extract <directory>
-```
-
-| Argument name(s) | Required | Default value                    | Description                |
-| :--------------- | :------- | :------------------------------- | :------------------------- |
-| directory        | Y        | NA                               | The name of the database   |
-| --basecalller    | N        | ONT Albacore Sequencing Software | The name of the basecaller |
-
----
-
-## filter
-
-### Overview
-
-Filter FASTQ files by length
-
-### Input
-
-- unfiltered reads
-
-### Output
-
-- filtered reads
-
-### Usage example
-
-```bash
-artic filter --max-length 500 --min-length 50 <filename>
-```
-
-| Argument name(s) | Required | Default value | Description                            |
-| :--------------- | :------- | :------------ | :------------------------------------- |
-| filename         | Y        | NA            | The reads to filter                    |
-| --max-length     | N        | NA            | Remove reads greater than max-length   |
-| --min-length     | N        | NA            | Remove reads less than than min-length |
-
----
-
 ## guppyplex
 
 ### Overview
@@ -182,14 +68,16 @@ artic minion <scheme> <sample>
 | :------------------- | :------- | :------------- | :------------------------------------------------------------------------------------------- |
 | scheme               | Y        | NA             | The name of the primer scheme                                                                |
 | sample               | Y        | NA             | The name of the sample                                                                       |
-| --clair3             | N        | False          | Use clair3 instead of medaka for variants  (experimental feature from v1.4.0)                |
 | --model       | Y        | NA             | Medaka or Clair3 model to use                                                                       |
 | --normalise          | N        | 100            | Normalise down to moderate coverage to save runtime                                          |
 | --threads            | N        | 8              | Number of threads                                                                            |
 | --scheme-directory   | N        | /artic/schemes | Default scheme directory                                                                     |
-| --max-haplotypes     | N        | 1000000        | Max-haplotypes value for nanopolish                                                          |
+| --scheme-name     | N        |                | Name of scheme to fetch from the primerschemes repository     |
+| --scheme-length     | N        |                | Length of scheme to fetch from the primerschemes repository     |
+| --scheme-version     | N        |                | Version of the scheme to fetch from the primerschemes repository     |
+| --bed                | N        |                | Bed file path            |
+| --ref                | N        |                | Reference fasta path     |
 | --read-file          | N        | NA             | Use alternative FASTA/FASTQ file to <sample>.fasta                                           |
-| --no-longshot        | N        | False          | Use medaka variant instead of longshot (experimental feautre from v1.2.0)                    |
 | --min-mapq           | Y        | 20             | Remove reads which map to the reference with a lower mapping quality than this               |
 | --no-indels          | N        | False          | Ignore insertions and deletions during variant calling, maintains the co-ordinates of the ref|
 | --no-frameshifts     | N        | False          | Do not allow frameshift variants (indels of lengths which are non divisible be 3 ) to be added to the consensus |
diff --git a/docs/faq.md b/docs/faq.md
index 25cc14a9..4a02bb6b 100644
--- a/docs/faq.md
+++ b/docs/faq.md
@@ -4,18 +4,19 @@ summary: The FAQ.
 authors:
   - Will Rowe
   - Nick Loman
+  - Sam Wilkinson
 date: 2020-03-30
 ---
 
 # FAQ
 
-## Where can I find the SOP for SARS-CoV-2
+## How do I process MPXV data?
 
-The standard operating proceedure for the ARTIC Network SARS-SoV-2 bioinformatics can be found [here](https://artic.network/ncov-2019/ncov2019-bioinformatics-sop.html).
+A set of resources for processing MPXV sequencing data may be found [here](https://artic.network/mpxv), this includes running this pipeline on the command line and the artic MPXV nextflow pipelines via epi2me.
 
-## Should I use the medaka or clair3 workflow
+## Where can I find the SOP for SARS-CoV-2?
 
-We currently recommend the medaka workflow as we have spent more time validating and supporting this workflow. That being said, both tend to give consistent results with our test datasets so the choice is yours.
+The standard operating proceedure for the ARTIC Network SARS-SoV-2 bioinformatics can be found [here](https://artic.network/ncov-2019/ncov2019-bioinformatics-sop.html).
 
 ## Lab-on-an-SSD
 
diff --git a/docs/installation.md b/docs/installation.md
index ce41f45d..b052288c 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -4,12 +4,13 @@ summary: The installation guide.
 authors:
   - Will Rowe
   - Nick Loman
+  - Sam Wilkinson
 date: 2020-03-30
 ---
 
 # Installation
 
-As of [release 1.4.0](https://github.com/artic-network/fieldbioinformatics/releases/tag/1.4.0), conda installation of fieldbioinformatics will become difficult due to the mutually exclusive requirements of medaka and clair3, for this reason we recommend either utilising the docker image [available here](https://quay.io/repository/artic/fieldbioinformatics) or to build the package from source after installing the dependencies via Conda.
+As of [release 1.4.0](https://github.com/artic-network/fieldbioinformatics/releases/tag/1.4.0), we provide a docker image [available here](https://quay.io/repository/artic/fieldbioinformatics) and a conda package. You may also wish to install the package from source after installing the dependencies via Conda yourself.
 
 ## Via conda
 
@@ -44,11 +45,10 @@ First check the pipeline can be called:
 artic -v
 ```
 
-To check that you have all the required dependencies, you can try the pipeline tests with both workflows:
+To check that you have all the required dependencies, you can try the pipeline tests like so:
 
 ```
 ./test-runner.sh clair3
-./test-runner.sh medaka
 ```
 
 For further tests, such as the variant validation tests, see [here](http://artic.readthedocs.io/en/latest/tests?badge=latest).
\ No newline at end of file
diff --git a/docs/minion.md b/docs/minion.md
index 37060d93..d2d78b15 100644
--- a/docs/minion.md
+++ b/docs/minion.md
@@ -4,7 +4,8 @@ summary: Outline of minion workflows.
 authors:
   - Will Rowe
   - Nick Loman
-date: 2020-09-01
+  - Sam Wilkinson
+date: 2024-11-11
 ---
 
 # Core pipeline
@@ -13,29 +14,20 @@ date: 2020-09-01
 
 This page describes the core pipeline which is run via the `artic minion` command.
 
-There are **2 workflows** baked into the core pipeline, one which uses signal data (via [nanopolish](https://github.com/jts/nanopolish)) and one that does not (via [medaka](https://github.com/nanoporetech/medaka)). As the workflows are identical in many ways, this page will describe the pipeline as whole and notify the reader when there is dfferent behaviour between the two workflows.
-It should be noted here that by default the `nanopolish` workflow is selected; you need to specify `--medaka` (and `--medaka-model`) if you want the medaka workflow enabled.
-
-> **NOTE**: It is very important that you select the appropriate value for `--medaka-model`.
-
 At the end of each stage, we list here the "useful" stage output files which are kept. There will also be some additional files leftover at the end of the pipeline but these can be ignored (and are hopefully quite intuitively named).
 
 ## Stages
 
 ### Input validation
 
-The supplied `--scheme-directory` will be checked for a **reference sequence** and **primer scheme**. If `--scheme-version` is not supplied, version 1 will be assumed.
+As of version 1.2.0, the pipeline will try and download the reference and scheme from the [artic primer scheme](https://github.com/artic-network/primer-schemes) repository if it is not found/provided. It will be downloaded to the directory provided by `--scheme-directory` which defaults to a subfolder of the current working directory `primer-schemes/`
 
-As of version 1.2.0, the pipeline will try and download the reference and scheme from the [artic primer scheme](https://github.com/artic-network/primer-schemes) repository if it is not found/provided.
+This is done using the `--scheme-name` `--scheme-length` and `--scheme-version` parameters, for example, to fetch the [artic-inrb-mpox/2500/v1.0.0](https://github.com/quick-lab/primerschemes/tree/main/primerschemes/artic-inrb-mpox/2500/v1.0.0) scheme you should provide the following arguments:
+* `--scheme-name artic-inrb-mpox`
+* `--scheme-length 2500`
+* `--scheme-version v1.0.0`
 
-Once the scheme and reference have been found, the pipeline will validate the scheme and extract the primer pool information.
-
-If running the **nanopolish workflow**, the pipeline will check the reference only contains one sequence and then run `nanopolish index` to map basecalled reads to the signal data.
-
-#### stage output
-
-- primer validation log (optional)
-- nanopolish index (workflow dependant)
+Alternatively you may provide the `--bed` and `--ref` arguments to point directly towards the primer bed file and reference fasta you wish to use.
 
 ### Reference alignment and post-processing
 
@@ -69,26 +61,17 @@ More information on how the primer scheme is used to infer amplicons can be foun
 
 ### Variant calling
 
-Once alignments have been softmasked, sorted and indexed (again), they are used for variant calling. This is where the two workflows actually differ.
-
-For the **medaka workflow**, we use the following commands on the `$SAMPLE.primertrimmed.rg.sorted.bam` alignment:
+We use the following commands on the `$SAMPLE.primertrimmed.rg.sorted.bam` alignment:
 
-- [medaka consensus](https://github.com/nanoporetech/medaka)
-- medaka variant or snps (if pipeline has been told not to detect INDELS via `--no-indels`)
-- medaka tools annotate (if `--no-longshot` has been selected)
-- [longshot](https://github.com/pjedge/longshot) (if `--no-longshot` not selected)
+- [run_clair3.sh](https://github.com/HKU-BAL/Clair3)
 
-And for the **nanopolish workflow** we use the following command on the `$SAMPLE.trimmed.rg.sorted.bam` alignment:
-
-- [nanopolish variants](https://github.com/jts/nanopolish)
-
-For both workflows, the variant calling steps are run for each **read group** in turn. We then merge variants reported per read group into a single file using the `artic_vcf_merge` module.
-
-> Note: we use the `$SAMPLE.trimmed.rg.sorted.bam` alignment for the **nanopolish workflow** as nanopolish requires some leading sequence to make variant calls; we use the primer sequence for this purpose in order to call variants at the start of amplicons.
+The variant calling steps are run for each **read group** in turn. We then merge variants reported per read group into a single file using the `artic_vcf_merge` module.
 
 Opionally, we can check the merged variant file against the primer scheme. This will allow us to detect variants called in primer and amplicon overlap regions.
 
-Finally, we use the `artic_vcf_filter` module to filter the merged variant file through a set of workflow specific checks and assign all variants as either PASS or FAIL. The final PASS file is subsequently indexed ready for the next stage.
+We then use the `artic_vcf_filter` module to filter the merged variant file through a set of workflow specific checks and assign all variants as either PASS or FAIL. The final PASS file is subsequently indexed ready for the next stage.
+
+Finally the variants which have passed filtering are normalised against the depth masked pre-consensus using [bcftools norm](https://github.com/samtools/bcftools) to ensure that the REF column in the VCF matches the pre-consensus for the final output.
 
 #### stage output
 
@@ -99,14 +82,15 @@ Finally, we use the `artic_vcf_filter` module to filter the merged variant file
 | `$SAMPLE.vcfreport.txt`  | a report evaluating reported variants against the primer scheme |
 | `$SAMPLE.fail.vcf`       | variants deemed too low quality                                 |
 | `$SAMPLE.pass.vcf.gz`    | detected variants (indexed)                                     |
+| `$SAMPLE.normalised.vcf.gz` | normalised variants (indexed)                                     |
 
 ### Consensus building
 
 Prior to building a consensus, we use the post-processed alignment from the previous step to check each position of the reference sequence for sample coverage. Any poition that is not covered by at least 20 reads from either read group are marked as low coverage. We use the `artic_make_depth_mask` module for this, which produces coverage information for each read group and also produces a coverage mask to tell us which coordinates in the reference sequence failed the coverage threshold.
 
-Next, to build a consensus sequence for a sample, we require a pre-consensus sequence based on the input reference sequence. The preconsensus has low quality sites masked out with `N`'s using the coverage mask and the `$SAMPLE.fail.vcf` file. We then use `bcftools consensus` to combine the preconsensus with the `$SAMPLE.pass.vcf` variants to produce a consensus sequence for the sample. The consensus sequence has the artic workflow written to its header.
+Next, to build a consensus sequence for a sample, we require a pre-consensus sequence based on the input reference sequence. The preconsensus has low quality sites masked out with `N`'s using the coverage mask and the `$SAMPLE.fail.vcf` file. We then use `bcftools consensus` to combine the preconsensus with the `$SAMPLE.normalised.vcf.gz` variants to produce a consensus sequence for the sample. The consensus sequence has the artic workflow written to its header.
 
-Finally, the consensus sequence is aligned against the reference sequence using `muscle`.
+Finally, the consensus sequence is aligned against the reference sequence using `muscle` if the `--use-muscle` parameter is provided.
 
 #### stage output
 
@@ -125,12 +109,4 @@ Finally, the consensus sequence is aligned against the reference sequence using
 | artic_vcf_filter          | filters a combined VCF into PASS and FAIL variant files                                              |
 | artic_make_depth_mask     | create a coverage mask from the post-processed alignment                                             |
 | artic_mask                | combines the reference sequence, FAIL variants and coverage mask to produce a pre-consensus sequence |
-| artic_fasta_header        | applies the artic workflow and identifier to the consensus sequence header                           |
-
-## Optional pipeline report
-
-As of version 1.2.1, if you run the pipeline with `--strict`, you can run MultiQC (which should be installed as part of the artic conda environment) on the pipeline output directory and this will produce a report containing amplicon coverage plots and variant call information. To generate a report from within your pipeline output directory:
-
-```
-multiqc .
-```
\ No newline at end of file
+| artic_fasta_header        | applies the artic workflow and identifier to the consensus sequence header                           |
\ No newline at end of file
diff --git a/docs/primer-schemes.md b/docs/primer-schemes.md
index 304cc42b..010bf67f 100644
--- a/docs/primer-schemes.md
+++ b/docs/primer-schemes.md
@@ -22,28 +22,19 @@ Tiling amplicon schemes contain primers that are designed using a greedy algorit
 
 ## Availability
 
-Frequently used primer schemes for viral genome sequencing are found in the [ARTIC primer scheme repo](https://github.com/artic-network/primer-schemes). These include Nipah, Ebola and SARS-CoV-2 primer schemes. Multiple scheme versions may be available, with a higher version number denoting schemes that have had improvements made to them (e.g. introduction of alternate primers).
-
-ARTIC primer schemes can be downloaded using the [artic-tools get_scheme](./commands.md#get_scheme) command. For example:
-
-```sh
-artic-tools get_scheme ebola --schemeVersion 2
-```
-
-The `artic pipeline` will also attempt to download a scheme using `artic tools` if a scheme can not be found locally.
+Frequently used primer schemes for viral genome sequencing are found in the [Quick-lab primer scheme repo](https://github.com/quick-lab/primerschemes).
 
 ## File format
 
 Primer schemes produced by [Primal Scheme](https://primalscheme.com/) come with several files (which you can read more about [here](https://github.com/aresti/primalscheme#output)). The 2 files which are needed for `artic` are:
 
-| file extension      | description                                                                    |
+| filename      | description                                                                    |
 | ------------------- | ------------------------------------------------------------------------------ |
-| `*.reference.fasta` | the sequence of the reference genome used in the scheme                        |
-| `*.primer.bed`      | the coordinates of each primer in the scheme, relative to the reference genome |
+| `reference.fasta` | the sequence of the reference genome used in the scheme                        |
+| `primer.bed`      | the coordinates of each primer in the scheme, relative to the reference genome |
 
-**Note**: `*.primer.bed` was formerly `*.scheme.bed`, but has lately changed to reflect a slightly updated file format. The old `*.scheme.bed` should still work fine and is available alongside the new versions at the [ARTIC primer scheme repo](https://github.com/artic-network/primer-schemes).
 
-The `*.primer.bed` file is in 6-column BED format, with the following column descriptions:
+The `primer.bed` file is in 7-column BED format, with the following column descriptions:
 
 | column | name       | type   | description                                               |
 | ------ | ---------- | ------ | --------------------------------------------------------- |
@@ -53,6 +44,7 @@ The `*.primer.bed` file is in 6-column BED format, with the following column des
 | 4      | name       | string | primer name                                               |
 | 5      | primerPool | int    | primer pool<sup>\*</sup>                                  |
 | 6      | strand     | string | primer direction (+/-)                                    |
+| 7      | primer     | string | the primer sequence                                       |
 
 <sup>\*</sup> column 5 in the BED spec is an int for score, whereas here we are using it to denote primerPool.
 
@@ -60,13 +52,13 @@ The `*.primer.bed` file is in 6-column BED format, with the following column des
 
 ## Scheme logic
 
-The following sections will discuss how primer schemes are read from a `*.primer.bed` file and used in the ARTIC software.
+The following sections will discuss how primer schemes are read from a `primer.bed` file and used in the ARTIC software.
 
 ### Reading schemes
 
 #### primers
 
-Each line in a `*.primer.bed` file is a single primer. Lines are processed one at a time and the input file does not need to be sorted in any way beforehand (although most schemes are sorted by primer start coordinate). As lines are read, they are converted to primer objects.
+Each line in a `primer.bed` file is a single primer. Lines are processed one at a time and the input file does not need to be sorted in any way beforehand (although most schemes are sorted by primer start coordinate). As lines are read, they are converted to primer objects.
 
 Most of the logic behind creating a primer object relies on the primer name (column 4). As a primer line is read, we check for the following tags in the name field:
 
@@ -74,16 +66,14 @@ Most of the logic behind creating a primer object relies on the primer name (col
 | -------- | -------------------- |
 | `_LEFT`  | the left primer      |
 | `_RIGHT` | the right primer     |
-| `_alt`   | the primer is an alt |
 
 **Important**:
 
 - tags are case sensitive
 - tags can be anywhere in the primer name but typically are placed at the end (e.g. REGION_42_RIGHT_alt)
 - a `_LEFT` or `_RIGHT` tag is required and only one is allowed per primer
-- the `_alt` tag is optional and denotes that the primer is an alternate version of another primer
 
-To merge a primer with its alternate into a single primer object, the two primers are combined such that a maximal span is achieved. The merged primer will have `_alt` dropped from the ID.
+To merge a primer with alternates into a single primer object, the two primers are combined such that a maximal span is achieved.
 
 Depending on the presence of the `_LEFT` or `_RIGHT` tag, the primer is assigned a direction (+/- to denote forward/reverse). Direction is also encoded in column 6 of newer schemes.
 
@@ -102,25 +92,7 @@ We can also call several helper methods on the object for getting things such as
 
 #### schemes
 
-Once all lines in a `*.primer.bed` file have been read and converted to primer objects they are held in a `scheme`. In `artic-tools`, this consists of 2 unordered maps, one for forward primers and one for reverse primers, as well as some additional datastructures to allow validation, index and query of schemes.
-
-### Validating schemes
-
-> Note: the following sections apply to `artic-tools`
-
-On reading the scheme file, it will have satisified the following checks:
-
-- scheme file must exist and be readable
-- scheme file must have at least 5 columns (tab separated)
-- each row must encode a primer (problem rows are flagged and validation fails after all rows are tried)
-- scheme file must not contain multiple reference sequence IDs (first column)
-
-The scheme data structure will then be validated:
-
-- the scheme must contain primers
-- the number of forward primers must match the number of reverse primers (this is **after** merging alts)
-- each forward primer must have a reverse primer with a matching base name within the same primer pool
-- no gaps must be present within the scheme (i.e. regions of the reference not covered)
+Once all lines in a `primer.bed` file have been read and converted to primer objects they are held in a `scheme`. In `artic-tools`, this consists of 2 unordered maps, one for forward primers and one for reverse primers, as well as some additional datastructures to allow validation, index and query of schemes.
 
 ### Querying schemes
 
diff --git a/docs/quick-start.md b/docs/quick-start.md
index 0e0bce01..b685e238 100644
--- a/docs/quick-start.md
+++ b/docs/quick-start.md
@@ -4,9 +4,10 @@ summary: A quick start guide for using the artic pipeline.
 authors:
   - Will Rowe
   - Nick Loman
-date: 2020-03-30
+  - Sam Wilkinson
+date: 2024-11-11
 ---
 
 # Quickstart
 
-This will be the quick start guide, for now it is best to use the [SOP on the artic website](https://artic.network/ncov-2019/ncov2019-bioinformatics-sop.html)
\ No newline at end of file
+This will be the quick start guide, for now it is best to use the [SOP on the artic website](https://artic.network/mpxv/mpxv-bioinformatics-sop.html)
\ No newline at end of file
diff --git a/docs/release-notes.md b/docs/release-notes.md
index e3f1b64d..58177ec2 100644
--- a/docs/release-notes.md
+++ b/docs/release-notes.md
@@ -13,69 +13,83 @@ date: 2020-03-30
 
 # Release notes
 
-## v1.1.0-rc1:
-* Support for read groups:
-    * Support ‘pool’ read groups taken from BED file, e.g.:
-        * nCoV2019_1
-        * nCoV2019_2
-    * These can be helpfully viewed in the Tablet viewer.
-    * Do variant calling separately on each group, permits detection of incongruent mutations e.g. from contamination
-* Update to nanopolish 0.12.5 fo more informative VCF output
-* Change nanopore filtering to support new VCF output (e.g. strand-bias)
-* Support indels with bcftools consensus
-* Longshot added to Medaka to permit filter VCF on depth
-* New workflow model - all commands run "one per sample", script nanopolish index
-* Add 'artic gupplyplex' for support for demultiplexed Guppy output to replace 'artic gather'
-* Fix for align_trim to remove erroneous CIGAR strains that can cause medaka to fail
-* Support for amplicons v3 format file (Will Rowe)
-* Test suite (Will Rowe)
-* Documentation (Will Rowe)
-* Remove old deprecated code (Will Rowe)
-
-## v0.13:
-* ARTIC fork
-
-## v0.12: 
-* Update to nanopolish v0.8.4 for Albacore 2.0+ support:
-    * run nanopolish -d /path/to/fast5 input.fasta before zibra.py minion
-
-## v0.11-dev:
-* Add stringent pipeline
-
-## v0.11 (25th July 2017):
-* Revert back to Ryan Wick's porechop repo as --untrimmed now implemented
-* New script: align_trim_fasta.py and reconstitute.py as basis for more principled demultiplexing (align first, then trim to known amplicon end points +/- 40 bases) to help prevent chimeric reads resulting in incorrect barcode identification
-
-## v0.10 (15th July 2017):
-* Better handling of confident deletions (previously ignored, now N-masked)
-* Update to nanopolish HEAD
-* Fixes to support latest nanopolish VCF format changes
-* Ability to choose higher values of max-haplotypes to support more divergent references
-
-## v0.9:
-* New command line interface zibra.py to ease extraction and demultiplexing with Albacore and multiple groups
-
-## v0.8:
-* Track latest Porechop with better adaptor trimming
-
-## v0.7:
-* Add Illumina wrapper scripts
-
-## v0.6:
-* Update poretools
-
-## v0.5:
-* Update nanopolish to latest HEAD - to fix Albacore 1.1 support (7de633d01cc35a58e5537af5dd1024ae0040d15c)
-* Add Lassa virus scheme
-
-## v0.4:
-* Update nanopolish to latest HEAD - aaaf90beb4efe37243c651e62fcfd11374fe2176 for Albacore 1 support
-* Initial support for Yellow Fever (500 and 1000 bp schemes)
-    * Move to poretools for extraction with named basecaller
-
-## v0.3:
-* add Porechop
-
-## v0.2:
-* update nanopolish to 0.6 - adds default support for various nanopore models
-* update samtools to 1.4
+v1.4.5:
+    * Fieldbioinformatics now supports rapid barcoded (fragmented) primer trimming and normalisation
+    * Nanopolish has been removed completely due to several compatibility issues
+    * Medaka has also been removed completely due to kicking out long indels in a way that cannot be changed.
+    * Clair3 is now the default variant caller, by default only the r9.4.1 models are available but a artic_get_models command has been added which will fetch the ONT created r10.4.1 models listed in the rerio repository.
+    * The pipeline will also attempt to pick an appropriate model based on the basecall_model_version_id field that is added to read headers by default by ONT sequencers.
+    * Removed longshot entirely, it also kicks out long variants and is now unnecessary due to clair3 being a much better variant caller.
+    * Primer scheme fetcher has been updated to pull from the quick-lab primal hub schemes repository. For schemes not available in this repository you may provide them directly with the arguments --bed and --ref.
+    * Automated docker builds pushing to quay.io for use in nextflow pipelines etc.
+    * Remove some old functionality which is no longer relevant (basecalling, gather, etc)
+    * Re-implement CI as a gh action.
+    * Fix the overlapping variants issue by normalising variants against the pre-consensus using bcftools norm.
+
+v1.1.0-rc1:
+    * Support for read groups:
+        * Support ‘pool’ read groups taken from BED file, e.g.:
+            * nCoV2019_1
+            * nCoV2019_2
+        * These can be helpfully viewed in the Tablet viewer.
+        * Do variant calling separately on each group, permits detection of incongruent mutations e.g. from contamination
+    * Update to nanopolish 0.12.5 fo more informative VCF output
+    * Change nanopore filtering to support new VCF output (e.g. strand-bias)
+    * Support indels with bcftools consensus
+    * Longshot added to Medaka to permit filter VCF on depth
+    * New workflow model - all commands run "one per sample", script nanopolish index
+    * Add 'artic gupplyplex' for support for demultiplexed Guppy output to replace 'artic gather'
+    * Fix for align_trim to remove erroneous CIGAR strains that can cause medaka to fail
+    * Support for amplicons v3 format file (Will Rowe)
+    * Test suite (Will Rowe)
+    * Documentation (Will Rowe)
+    * Remove old deprecated code (Will Rowe)
+
+v0.13:
+    * ARTIC fork
+
+v0.12: 
+    * Update to nanopolish v0.8.4 for Albacore 2.0+ support:
+        * run nanopolish -d /path/to/fast5 input.fasta before zibra.py minion
+
+v0.11-dev:
+    * Add stringent pipeline
+
+v0.11 (25th July 2017):
+    * Revert back to Ryan Wick's porechop repo as --untrimmed now implemented
+    * New script: align_trim_fasta.py and reconstitute.py as basis for more principled demultiplexing (align first, then trim to known amplicon end points +/- 40 bases) to help prevent chimeric reads resulting in incorrect barcode identification
+
+v0.10 (15th July 2017):
+    * Better handling of confident deletions (previously ignored, now N-masked)
+    * Update to nanopolish HEAD
+    * Fixes to support latest nanopolish VCF format changes
+    * Ability to choose higher values of max-haplotypes to support more divergent references
+
+v0.9:
+    * New command line interface zibra.py to ease extraction and demultiplexing with Albacore and multiple groups
+
+v0.8:
+    * Track latest Porechop with better adaptor trimming
+
+v0.7:
+    * Add Illumina wrapper scripts
+
+v0.6:
+    * Update poretools
+
+v0.5:
+    * Update nanopolish to latest HEAD - to fix Albacore 1.1 support (7de633d01cc35a58e5537af5dd1024ae0040d15c)
+    * Add Lassa virus scheme
+
+v0.4:
+    * Update nanopolish to latest HEAD - aaaf90beb4efe37243c651e62fcfd11374fe2176 for Albacore 1 support
+    * Initial support for Yellow Fever (500 and 1000 bp schemes)
+        * Move to poretools for extraction with named basecaller
+
+v0.3:
+    * add Porechop
+
+v0.2:
+    * update nanopolish to 0.6 - adds default support for various nanopore models
+    * update samtools to 1.4
+
diff --git a/docs/tests.md b/docs/tests.md
index 136b20ee..2ea20d84 100644
--- a/docs/tests.md
+++ b/docs/tests.md
@@ -4,12 +4,13 @@ summary: Description of the available tests
 authors:
   - Will Rowe
   - Nick Loman
-date: 2020-03-30
+  - Sam Wilkinson
+date: 2024-11-11
 ---
 
 # The test suite
 
-All of the `artic` tests are run as part of the [Travis Continuous Integration](https://travis-ci.org/github/artic-network/fieldbioinformatics) testing. To run any of the tests yourself, it is assumed that you have downloaded the codebase and are using an appropiate environment:
+All of the `artic` tests are run as part of the as github actions. To run any of the tests yourself, it is assumed that you have downloaded the codebase and are using an appropiate environment:
 
 ```
 git clone https://github.com/artic-network/fieldbioinformatics
@@ -31,15 +32,14 @@ pytest -s artic/*_unit_test.py
 To test the core pipeline, you can use the `test-runner.sh` bash script. You can test both the medaka and nanopolish workflows:
 
 ```
-./test-runner.sh medaka
-./test-runner.sh nanopolish
+./test-runner.sh clair3
 ```
 
 The pipeline tests use a small subset of an Ebola virus amplicon sequencing run (flongle) which is downloaded from [here](http://artic.s3.climb.ac.uk/run-folders/EBOV_Amplicons_flongle.tar.gz) when the test is called.
 
 ## Variant validation tests
 
-Finally, we have also included some validation tests that will download several reference SARS-CoV-2 datasets, run the nanopolish/medaka workflows and then validate the reported variants and consensus sequences. To run all of the available validation datasets:
+Finally, we have also included some validation tests that will download several reference SARS-CoV-2 datasets, run the workflow, and then validate the reported variants and consensus sequences. To run all of the available validation datasets:
 
 ```
 pytest -s artic/minion_validator.py