change log and readme for releasing 1.2.6

CHANGELOG.md

@@ -1,3 +1,10 @@
+### 1.2.6 / 2019-07-13
+-   Refactored and fixed a bug in local assembler
+-   Refactored `megahit` script
+-   Obtain total memory size from `os.sysconf`
+-   Fixed segmentation fault in Mac OS with clang 4.0
+-   Added `--cleaning-rounds` and `--disconnect-ratio` options for more flexible graph cleaning control
+
 ### 1.2.5-beta / 2019-06-28 PST
 -   Fixed a bug that causes higher memory usage in seq2sdbg
 -   Refactor on sequence sorters, edge I/O and contig I/O

Dockerfile

@@ -8,6 +8,7 @@ RUN mkdir -p build
 WORKDIR build
 RUN cmake -DCMAKE_BUILD_TYPE=Release ..
 RUN make -j4
+RUN make simple_test
 RUN make install
-RUN megahit --test --no-hw-accel
-RUN megahit --test --no-hw-accel --kmin-1pass
+RUN megahit --test
+RUN megahit --test --kmin-1pass

README.md

@@ -5,48 +5,34 @@ MEGAHIT
 
 MEGAHIT is an ultra-fast and memory-efficient NGS assembler. It is optimized for metagenomes, but also works well on generic single genome assembly (small or mammalian size) and single-cell assembly.
 
-*News: try v1.2.x!*
-------
-
-MEGAHIT v1.2.x (beta) is released. Compared to v1.1.x, its changes include
-
--   faster and more memory-efficient than before, by using [BMI2 instructions](https://en.wikipedia.org/wiki/Bit_Manipulation_Instruction_Sets) and more recent hash libraries ([phmap](https://github.com/greg7mdp/parallel-hashmap) and [xxhash](https://github.com/Cyan4973/xxHash))
--   refactored with C++11 features
--   use CMake to build the project
--   removal of GPU support
-
-Please follow the instructions in [Getting Started](#gst) to try this new version.
-Past versions can be found at the [release](https://github.com/voutcn/megahit/releases) page.
-
-<a name="gst"></a>Getting Started
+Installation
 ---------------
 
-### Running with Linux binaries or docker images (recommended)
+#### Prebuilt binaries for x86_64 Linux
 
-``` sh
-wget https://github.com/voutcn/megahit/releases/download/v1.2.5-beta/MEGAHIT-1.2.5-beta-Linux-static.tar.gz
-tar zvxf MEGAHIT-1.2.5-beta-Linux-static.tar.gz
-cd MEGAHIT-1.2.5-beta-Linux-static/bin/
+```sh
+wget https://github.com/voutcn/megahit/releases/download/v1.2.6/MEGAHIT-1.2.6-Linux-x86_64-static.tar.gz
+tar zvxf MEGAHIT-1.2.6-Linux-x86_64-static.tar.gz
+cd MEGAHIT-1.2.6-Linux-x86_64-static/bin/
 ./megahit --test  # run on a toy dataset
-./megahit -1 YOUR_PE_READ_1.gz -2 YOUR_PE_READ_2.fq.gz -o YOUR_OUTPUT_DIR
+./megahit -1 MY_PE_READ_1.fq.gz -2 MY_PE_READ_2.fq.gz -o MY_OUTPUT_DIR
 ```
 
-You can also run MEGAHIT with its docker images.
-
+#### Prebuilt Docker image
 ``` sh
 # in the directory with your input reads
 docker run -v $(pwd):/workspace -w /workspace --user $(id -u):$(id -g) vout/megahit \
-  megahit -1 YOUR_PE_READ_1.gz -2 YOUR_PE_READ_2.fq.gz -o YOUR_OUTPUT_DIR
+  megahit -1 MY_PE_READ_1.fq.gz -2 MY_PE_READ_2.fq.gz -o MY_OUTPUT_DIR
 ```
 
-### Building from source
+#### Building from source
 
-#### Prerequisites
+##### Prerequisites
 
 -   For building: zlib, cmake &gt;= 2.8, g++ &gt;= 4.8.4
 -   For running: gzip and bzip2
 
-``` sh
+```sh
 git clone https://github.com/voutcn/megahit.git
 cd megahit
 git submodule update --init
@@ -60,19 +46,22 @@ make simple_test  # will test MEGAHIT with a toy dataset
 Usage
 -----
 
-To run MEGAHIT with default parameters:
-
-``` sh
-megahit -1 YOUR_PE_READ_1.fq.gz -2 YOUR_PE_READ_2.fq.gz -r YOUR_SE_READ.fq.gz -o YOUR_OUTPUT_DIR
+#### Basic usage
+```sh
+megahit -1 pe_1.fq -2 pe_2.fq -o out  # 1 paired-end library
+megahit --12 interleaved.fq -o out # one paired & interleaved paired-end library
+megahit -1 a1.fq,b1.fq,c1.fq -2 a2.fq,b2.fq,c2.fq -r se1.fq,se2.fq -o out # 3 paired-end libraries + 2 SE libraries
+megahit_core contig2fastg 119 out/intermediate_contigs/k119.contig.fa > k119.fastg # get FASTG from the intermediate contigs of k=119
 ```
+The contigs can be found `final.contigs.fa` in the output directory.
 
-If you did not install Megahit to your PATH, just run Megahit with full-path, e.g.
-
-``` sh
-/PATH/TO/MEGAHIT/build/megahit
-```
+#### Advanced usage
+- `--kmin-1pass`: if sequencing depth is low and too much memory used when build the graph of k_min
+- `--presets meta-large`: if the metagenome is complex (i.e., bio-diversity is high, for example soil metagenomes)
+- `--cleaning-rounds 1 --disconnect-ratio 0`: get less pruned assembly (usually shorter contigs)
+- `--continue -o out`: resume an interrupted job from `out`
 
-To see the full manual of Megahit, run the program without parameters or with `-h`.
+To see the full manual, run `megahit` without parameters or with `-h`.
 
 Also, our [wiki](https://github.com/voutcn/megahit/wiki) may be helpful.